Poverty, policies and perceptions in Tanzania

iob evaluations | nr 296

Poverty, policies and perceptions in Tanzania

An evaluation of Dutch aid to two district rural development programmes

Policy and operations Evaluation Department | FEBRUARY 2004

iob evaluations | nr. 296Policy and Operations Evaluation Department | Ministry of Foreign Affairs

Poverty, policies and perceptions in Tanzania

An evaluation of Dutch aid to two district rural development programmes

II

Printing : Opmeer | De Bink | TDS printmaildata

Design : Annelies Glandorf | b.ont

Bianca Wesseling | Jean Cloos Art Direction

Cover photograph: Lineair

ISBN: 90-5328-337-4

Ordercode : OSDR 0517/E

www.euforic.org/iob

www.minbuza.nl

February 2004

Preface

The principal objective of development cooperation as practised by the Netherlands is the

reduction of poverty. The degree to which development activities contribute towards that

objective is the most difficult question that can be raised in evaluations of the ways in

which development policy is implemented, and forms the focal point of the present

research. The approach chosen is the way in which the people perceive their own poverty

situation (for which the activities are intended), as well as long-term social and economic

trends. These are then confronted with the manner in which development funds are

employed. The evaluation is oriented towards two districts of Tanzania – Mbulu and

Songea – where the Netherlands has made long-term financial contributions to District

Rural Development Programmes.

New methods and techniques were adopted in confronting perceptions, trends and policy

in the two districts. In Mbulu a video registration was made of the people’s perceptions,

but due to the costs involved this was not feasible in Songea. In both districts the people’s

views were recorded by a research team from the University of Dar es Salaam. Literature

with regard to social and economic trends was studied and summarised. It was no simple

matter to gather and assemble all elements, efficiently and effectively. Even after further

research, there were gaps in the database with regard to social and economic changes.

This was a reason why the Terms of Reference could not entirely be fulfilled. The extra

research unfortunately caused considerable delay in rounding-off the evaluation.

Nevertheless, the evaluation has provided valuable insight into the problems that are

linked to regional development activities, oriented towards poverty reduction.

The research has shown that poverty in the districts has not lessened. Enabling policies at

the national level have proven to offer an insufficient basis at the local level for the dis-

tricts to develop independently. In general, aid provided by the Netherlands has been

inadequate to make any real difference. In designing the form of that aid, two problems

occurred: insufficient orientation towards the real causes of poverty as seen by the popu-

lation, and frequent changes in the focus of the aid. Those changes were due in particular

to political re-orientation on the part of the Netherlands, rather than to growing insight

into the poverty situation in the districts concerned or into the opinions held by the local

people.

III

Poverty, policies and perceptions in Tanzania |

Preface

It is no simple matter to record the people’s perceptions in writing – it is far more effec-

tive to let them speak for themselves. That was the reason why a film was compiled from

the enormous amount of video material, which is attached to this report in CD-Rom for-

mat.

The research was carried out under the responsibility of Inspector Alex Bartelink of the

Policy and Operations Evaluation Department (IOB) of the Netherlands Ministry of

Foreign Affairs. Sjoerd Zanen and Jacoline Plomp were involved in the research as consul-

tants. Data were compiled in the districts by a group of Tanzanian researchers under the

leadership of C.S.L. Chachage. Ton Aarden was involved as filmer in making a video regis-

tration of the people’s views. When Inspector Bartelink retired on reaching pensionable

age, consultant Jan Sterkenburg took on the responsiblity of rounding-off the investiga-

tion, assisted by consultant Dick Bol on certain points in the reporting. Finally, a word of

thanks is due to all those people in Tanzania and the Netherlands who made the research

feasible, particularly the inhabitants of the districts who were prepared to speak with the

researchers regarding their own conditions of life.

Final reponsibility for this evaluation is naturally that of IOB.

Rob van den Berg

Director

Policy and Operations Evaluation Department February 2004


Preface

IV

CONTENTS

V


Contents

Preface III

Tables and Figures IX

Abbreviations XI

1 Main findings 11.1 Introduction 1

1.2 Main findings 2

1.3 Issues for future attention 5

2 Objectives and evaluation approach 72.1 Justification 7

2.2 Objectives and key questions 8

2.3 Scope of the study 9

2.4 Methodology 11

2.5 Structure of the report 12

3 Poverty, development and aid 133.1 Poverty and development 13

3.2 Poverty reduction and decentralisation 16

3.3 Poverty alleviation in Dutch development co-operation 18

3.4 Summary and conclusions 21

4 Changes in the economy and local governance in rural Tanzania 234.1 Introduction: rural Tanzania before Independence 23

4.2 Socialism and rural development 1961-1980 24

4.3 Economic liberalisation 1980-2002 26

4.4 Decentralisation and local governance 28


VI


Contents

5 Poverty and government policy in Tanzania 335.1 Poverty in Tanzania 33

5.2 Poverty and government policy 1961-2002 38


6 Dutch aid and poverty alleviation in Tanzania 456.1 Background 45

6.2 Poverty alleviation in Dutch assistance to Tanzania 47

6.3 District programmes as a tool for poverty alleviation 50


7 Economic change and local governance in Mbulu district 597.1 District profile 59

7.2 Economic change 62

7.3 Social change 64

7.4 Local governance 66

7.5 Perceptions and poverty 67


8 District rural development programme in Mbulu 738.1 Introduction 73

8.2 Objectives and strategy 73

8.3 Programme organisation and finances 75

8.4 Activities and output 79

8.5 Impact on poverty 82


9 Economic change and local governance in Songea district 899.1 District profile 89

9.2 Economic change 92

9.3 Social change 95

9.4 Local governance 97

9.5 Perceptions and poverty 99


10 District rural development programme in Songea 10510.1 Introduction 105

10.2 Objectives and strategy 105

10.3 Programme organisation and finances 107




11 Assessment of drdp’s contribution to poverty alleviation 12111.1 Socio-economic change in rural Tanzania 121

11.2 Poverty and perceptions of poverty 123

11.3 Local governance and poverty reduction 125

11.4 Principal characteristics of district development programmes 126



AnnexesAnnex 1 The Policy and Operations Evaluation Department 135

Annex 2 Terms of Reference location specific poverty study 137

Annex 3 Tables 145

Annex 4 References 165

Annex 5 Exchange rates 1986 –2001 171

VII


Contents


Contents

VIII

IX


Tables and Figures

Tables

5.1 Poverty changes during the 1990s 35

5.2 Trends in Social Indicators during the period 1960-2000 35

5.3 Changes in Social Indicators during the 1990s 36

5.4 Government expenditure in the social sectors TSH bn 40

7.1 Production of main food crops 1991-1997, including Karatu 64

8.1 MDRDP expenditure 1988-2002 76

8.2 Sector Distribution Domestic Expenditure MDRDP, 1988-1993 77

8.3 MDRDP sector distribution for development expenditure, 1997-2002 78

8.4 Trends in financial assistance, Mbulu DRDP, 1988-2002 79

9.1 Production and value of main crops, Songea District 1988-2000 95

10.1 Total expenditure SODA/DRP 1994-2002 108

10.2 SODA/DRDP - Sector distribution of development expenditure 1994-2002 109

10.3 Main categories of development expenditure SODA/DRDP 1994-2002 109

Annex 3.1 Expenditure on DRDP 1987-2000 145

Annex 3-2 Major Activities Mbulu DRDP 146

Annex 3-3 Major Output MDRDP during 1998-2000 148

Annex 3-4 Activities and output SODA/DRDP, 1994-2002 150

Annex 5 Exchange rates 1986 -2001 171

Figures

5.1 Per Capita Income Growth (1960-2000) 34

6.1 Aid to Tanzania 1970-1996 46

Maps

1 DRDP supported by the Netherlands 52

2 Mbulu District 61

3 Songea District 90

Tables and Figures


Tables and Figures

X

Abbreviations

AIV Advisory Council on International Affairs

ALAT Association of Local Authorities in Tanzania

ASDS Agricultural Sector Development Strategy

CBO Community Based Organisation

CCM Chama Cha Mapinduzi (formerly TANU)

DAC Development Assistance Committee

DADP District Agricultural Development Plan

DED District Executive Director

DFID Department for International Development (U.K.)

DRDP District Rural Development Programme

GDP Gross Domestic Product

HIPC Highly Indebted Poor Country

IDS Institute of Development Studies

IFAD International Fund for Agricultural Development

ILO International Labour Organisation

IOB Policy and Operations Evaluation Department (formerly IOV)

IOV Inspectie Ontwikkelingssamenwerking te Velde (Operations Review Unit - later

IOB)

IRDP Integrated Rural Development Programme

LGA Local Government Authority

LGRP Local Government Reform Programme

MDRDP Mbulu District Rural Development Project

NGO Non-Governmental Organisation

NGOMAT Ngoni-Matengo Co-operative Union

XI


Abbreviations

NMC National Milling Corporation

ODA Official Development Assistance

OECD Organisation for Economic Cooperation and Development

OED Operations Evaluation Department (World Bank)

PRSP Poverty Reduction Strategy Paper(s)

RCU Ruvuma Co-operative Union

RDU Ruvuma Development Association

RECO Regional Co-ordinator (of DRDP)

RNE Royal Netherlands Embassy

SAMCU Songea Agricultural Marketing Co-operative Union

SNV Netherlands Development Organisation/SNV

SODA Songea Development Action

TANU Tanzania African National Union (later CCM)

TAT Tobacco Authority of Tanzania

UN United Nations

UNDP United Nations Development Programme

UNICEF United Nations Children's Fund

WB World Bank

WDC Ward Development Committee


Abbreviations and Acronyms

XII

1 Main findings

1.1 Introduction

The objective of this evaluation basically was to examine the extent to which interventions

supported by the Netherlands in the framework of the District Rural Development

Programmes (DRDPs) in Tanzania have addressed the fundamental causes of rural pover-

ty as perceived by the various stakeholders in the programmes. DRDPs have poverty

reduction as their principal objective. They are conducted within a concrete geographical

context which makes it possible to describe the longer-term processes of rural change

that affected rural poverty. Dutch support for district development in Tanzania started in

1987 in four districts, gradually expanding to fourteen districts at the present time.

Over the period 1987-2002 expenditure totalled about US$ 60 million, equalling about

10% of Dutch aid to Tanzania during those years.

Two of the fourteen districts were selected for the study, i.e. Mbulu and Songea, based on

their peripheral location, their internal socio-economic differentiation and the protracted

period of support provided by the Netherlands, i.e. since 1987 in Mbulu and since 1993 in

Songea. A considerable difference in population density between the two districts is

reflected in land availability per capita; this influences the potential for agricultural

output growth and, thus also rural poverty conditions. The two districts are not represen-

tative of all fourteen districts, but each is representative of several others, i.e. those with

low potential for agricultural production and relatively high population pressure (Mbulu),

or with relatively good potential provided attention is given to maintaining soil fertility

(Songea).

Initially, the overall aim of DRDPs was to increase the income of farm households and to

improve the living conditions of the rural population. Later, this was re-formulated as: to

improve the well-being of the rural population in a sustainable way, thus reflecting a shift

in focus from support for income generation to the improvement of social services.

Recently, i.e. with the incorporation of DRDP into sector support to local governance,

the focus again shifted, towards the strengthening of local governance.

The evaluation was experimental in several respects: it attempted to reveal the effects

of long-term processes of change on poverty, and to relate the relevance of the aid

1


Main findings

programme to the people’s perception of poverty and its basic causes. Moreover, it used

video recordings as an instrument of data collection in addition to other methods.

The study had several limitations, the most important of which were the complications

involved in extrapolating the perceptions of categories of respondents at village level to

the poverty situation at district level, and the lack of reliable data on long-term trends in

agricultural production, income and social services. Despite these limitations, the evalu-

ation provides interesting insights into poverty reduction effects of Dutch aid as perceived

by the local population, elected leaders and government staff.

1.2 Main findings

2.1 District Rural Development Programmes have not achieved their main long term objective:

the improvement of living conditions and the structural alleviation of poverty.

The district programmes supported a wide-range of activities directed towards rehabilita-

tion of the physical infrastructure, strengthening of the economic base, improvement of

social services, and enhancement of human capabilities in local governance. Despite

10-15 years of support and considerable achievements in various fields of activity, the

impact on the poverty suffered by rural households has been limited. The two districts

experienced an increase rather than a decrease of poverty, a trend that corresponds

largely with that for Tanzania as a whole: national statistics indicate no real per capita

income growth over the past 25 years.

The low average volume of aid per capita, low coverage of households for income

generating activities (between 5 and 12%), the absence of any significant breakthrough in

agricultural productivity, and the location-specific improvements of social services, are

indicative of the limited impact on poverty alleviation. The Dutch contribution was frag-

mented, moreover, over a wide range of activities, whereas there was a need for strategic

interventions in key areas for poverty reduction in view of the low volume of aid in per

capita terms.

There are two principal causes for the disappointing impact on poverty. Firstly, the

refractory socio-political environment in Tanzania and the non-conducive government

policy. Second, deficiencies in the aid approach, including the non-identification of the

basic problems of the poor and several changes in DRDP focus during implementation.

Consequently, DRDP support was only inadequately targeted towards the poor, if at all.

If DRDPs continue in the present way, their contribution to poverty reduction will be

negligible also in future.


Main findings

2

2.2 The changes in DRDP focus were largely inspired by Dutch development co-operation policy,

and received a positive response from politicians and bureaucracy in Tanzania.

The emphasis in DRDP shifted from strengthening the economic base and agricultural

production towards improving social services and, subsequently, towards enhancing the

capabilities of local governance. These changes were largely inspired by Dutch develop-

ment co-operation policy. From the late 1980s onwards more attention was given to the

social sectors and to ownership of aid activities on the part of the recipient government

and, later, also to institutional development and good governance. The shift towards

capacity building in local governance, in particular, made strengthening the economic

base for increasing rural incomes into a secondary objective

Conditions in Tanzania for the observed changes were favourable: while economic

reforms reduced the role of the state in the development process, social services remained

largely in the public domain. Moreover, local politicians and civil servants welcomed the

support given to social services, including the rehabilitation and construction of infra-

structure, as tangible signs of progress, and profited from the capacity building efforts.

Ownership of the programme, therefore, became largely that of elected leaders,

politicians and civil servants, rather than of the village population

2.3 District Rural Development Programmes have given little attention to the basic obstacles to

poverty alleviation as perceived by the rural population; removal of some of these obstacles

was beyond the authority of the district administration and required intervention at the

national level.

In the perception of the villagers, those projects aimed at improving social services and

physical infrastructure were relevant for poverty alleviation. In their opinion, however, the

DRDP did not address their crucial problems: the lack of an enabling environment to

increase agricultural production and to market their produce at fair prices. The popula-

tion did not perceive the basic causes of poverty in terms of income, but rather in the

socio-political conditions that influenced production, income and welfare: access to and

control over land, the quality and reliability of support services for agricultural production

and animal husbandry (input supply, marketing, veterinary services), the lack of social

security due to the absence of employment opportunities, and weaknesses in leadership

and the legal system.

There were two principal reasons for not addressing these priorities as perceived by the

population. Firstly, the lack of any thorough analysis of the main obstacles to rural devel-

opment and the basic causes of rural poverty in the preparation phase of the district pro-

3


Main findings

grammes and, until recently, shortcomings in monitoring and evaluations which

prevented such obstacles becoming known.

Second, the limited possibilities to enable such crucial problems to be solved at the

district level. Several principal obstacles towards district development fall outside the

authority of district administrations and can only be addressed effectively at the national

level. The sector-wide approach which, among other things, is intended to link local level

problems to national level policy, and the incorporation of DRDPs in sector support to

local governance, did not create the appropriate framework in which effectively to address

problems resulting from insufficient marketing liberalisation and inadequately-function-

ing marketing structures. This is chiefly due to local governance not being a genuine

sector for channelling district level issues to relevant national policy institutions.

2.4 DRDPs have made little effort towards political empowerment and the strengthening of civil

society; those attempts that were made were largely ineffective.

The Netherlands started its support to district development in Tanzania in the mid-1980s

when the policy environment for such support was rather unfavourable: Tanzania did not

have a poverty alleviation strategy, nor a decentralisation policy that included delegation

of authority to the district level under a multi-party political system. Over time, the policy

environment has improved, but concrete changes at the local level have been implement-

ed only slowly. At the district level, Tanzania is de facto still a one-party state, with a

government administration linked to the party machinery. Perception studies reveal that,

at the village and ward level, many people either fear their leaders or state that leaders

have to be obeyed. Others, especially younger people, complain about the quality of

leadership, corruption, lack of transparency and mismanagement of funds.

The recent shift to local governance focused on capacity building of the district

administration and council staff rather than on empowerment of the local population.

And although the programmes introduced participatory planning methods, various types

of training and group formation, there was no systematic approach to the creation of

countervailing power (e.g. strategic support of NGOs and strengthening of village

councils). Consequently, at the village level the population saw participatory planning as

a bureaucratic exercise with all decisions being taken by government officials, and as a

technical method of project preparation and implementation, rather than as a means to

democratise village structures.


Main findings

4

Experiences elsewhere in Sub-Sahara Africa regarding the effectiveness of donor assis-

tance for decentralisation-cum-poverty alleviation, indicate the importance of combining

the organisational strengthening of local administrations with political empowerment of

the population and mobilisation of productive resources in a comprehensive strategy.

2.5 The orientation of aid towards resource-poor and remote regions as pockets of poverty reduces

the chances of effectively achieving poverty alleviation.

Although support given to Songea was of shorter duration, achievements were generally

better than in Mbulu, partly because of the better resource potential, ample availability of

land, and possibilities for technological innovation. This difference in achievements

between Mbulu and Songea confirms the importance of area selection for rural develop-

ment support.

In selecting project areas, a donor who gives priority to marginal areas is confronted with

a dilemma. As has been observed in several other evaluations, little can be achieved in

terms of sustainable development in such areas, in view of the limited possibilities for

economic growth, even under conditions of long periods of donor support and substan-

tial amounts of aid The natural resource situation constrains agricultural output and

surplus production, while the peripheral location restricts marketing opportunities.

In view of the limited possibilities for economic growth in marginal areas, there seems to

be good reason to give priority to those areas with moderate to good potential.

These may become foci of growth from which benefits spread to other areas or,

alternatively, which attract migrants from low potential areas.

1.3 Issues for future attention

If the Netherlands intends to continue its support to district development programmes

and effectively to address poverty as a consequence of the principal objective of its

development co-operation policy, the following issues require further attention:

3.1 A strategy for structural poverty alleviation in the wider framework of support to decentralised

rural development should be operationalised in greater detail. Such a strategy must include

identification of the resource potential and of the basic constraints to poverty reduction, and of

the approaches needed to make effective use of that potential and to remove the constraints to

the benefit of the poor. Democratisation and empowerment of the local population is also an

essential element of an effective poverty reduction strategy.

5


Main findings

3.2 A district-specific strategy for rural poverty alleviation requires specification of the issues that

can effectively be addressed at district level and of those that need measures at national level.

This would be part and parcel of a sector-wide approach that intends to link issues at the

micro-, meso-, and macro level. The incorporation of DRDPs in sector support for local

governance must also provide insight into how to address national-level issues outside the

immediate scope of local governance, such as marketing liberalisation in agriculture.

Moreover, the role of government and of the private sector with regard to poverty alleviation

should be clearly defined.

3.3 Any programme for poverty reduction in rural areas must indicate ways by which to increase

production, productivity and profitability of agriculture. Agricultural production growth faces

both technological and political/institutional constraints. Most aid programmes for rural

development emphasise the removal of technological constraints; political/institutional

constraints, however, are often of greater importance for effective poverty alleviation and

therefore need explicit attention in the design of such programmes.


Main findings

6

2 Objectives and evaluationapproach

2.1 Justification

Reducing poverty has been the major goal of development co-operation for the past

50 years. Remarkable results have been achieved in the economic and social sectors of

developing countries: during the last 30 years, life expectancy has increased by an

average of more than 20 years, infant mortality rates have dropped by half and primary

school enrolment rates have doubled. In the last century, average income, health and

literacy have improved at a faster pace in the developing countries than in the OECD

countries (OECD/DAC, 2000). Development co-operation may take part of the credit for a

number of these successes.

However, progress in terms of both income and social services has mainly taken place in

East and Southeast Asia, and improvements observed there do not apply to Africa south

of the Sahara. From 1990-2000, the number of people with an income of less than US$ 1

per day in Sub-Sahara Africa increased from 242 million to 291 million. This means that,

at the turn of the century, almost half the population of this continent was classed as

poor. The poverty situation is also reflected in nutritional data: according to UN statistics,

almost one third of the African population is undernourished. Bad governance, wars and

corruption are frequently mentioned as causes of the African deprivation, as well as such

other internal factors as a fragile environment and poor infrastructure. In addition, trade

restrictions introduced by rich countries to protect their own economies may also affect

countries in Sub-Sahara Africa.

Some 50 years ago Tanzania was one of the poorest countries in Sub-Sahara Africa and it

is still so today. Since the late 1960s its government has made efforts to address poverty,

as is apparent from the range of its policy documents. The country also received massive

volumes of foreign aid from numerous donors, most of whom saw poverty alleviation as

the principal aim of their assistance policies. Why did these efforts have such meagre

results?

7


Objectives and evaluation approach

Since the early 1970s, the reduction of poverty has also been the major objective of Dutch

development co-operation policy. For quite some time, improving the living conditions of

poor groups was chiefly addressed through projects, either through improvements in

community services such as education, health care and drinking water supply, or through

support for activities that generate income. In evaluating these projects, the emphasis

was placed usually on the degree to which the goals had been achieved and the efficiency

of implementation. Generally, evaluations started with an analysis of government policy

and goals of projects and programmes, rather than with an analysis of the poverty prob-

lem in the field. Evaluations also paid far less attention to (i) processes that led to this sit-

uation and the underlying causes of poverty, (ii) the perception of the population regard-

ing the poverty situation, its underlying causes and the relevance of aid activities, and (iii)

the extent to which aid activities aimed at removing the causes of poverty as perceived by

the population.

2.2 Objectives and key questions

The general objective of the present evaluation is to examine the extent to which interven-

tions financed by the Netherlands in the framework of the District Rural Development

Programme have addressed the basic causes of poverty as perceived by the local popula-

tion. This means that the perception of the rural population has been the starting point of

the study, but present-day perceptions are placed in the wider framework of long-term

processes of change and the significance of those changes for different segments of the

population. In addition, the study examined the extent to which the interventions

financed by the Netherlands were in line with the dynamics of change in rural areas, and

the perception and priorities of the population. This is considered to be an indication of

the relevance and potential effectiveness of the aid programme in reducing poverty and

improving the population’s living conditions

This general objective has been specified by means of the following key questions:

1. What processes can be identified in the rural areas that influence(d) changes in the living condi-

tions and degree of poverty of different segments of the population?

2. What were the main characteristics of government policy in Tanzania with regard to poverty

reduction and to what extent was this policy in line with the processes of change in rural areas and

the people’s perception of poverty?



8

3. What has been the contribution of Dutch development co-operation in the framework of the

district development programmes to the reduction of poverty, and how can these activities be

evaluated in relation to the poverty situation and the perception of the population?

The processes of change were analysed in selected areas against the background of

general changes in rural Tanzania after independence and in relation to the rural develop-

ment policy pursued by the Tanzanian government. The study concentrates on those

changes that influenced social and economic developments in rural areas on the one

hand and local governance on the other. These are expected to provide insight into the

background and causes of the current poverty situation.

In the context of the rural economy, this chiefly involves changes in farming and agricul-

tural services and the effects on agricultural production and rural incomes. In the social

sectors, the magnitude, distribution and performance of education, health and drinking

water facilities are considered indicative of living conditions of the rural population.

Local governance includes the improvement of district administration and democratisa-

tion, i.e. strengthening of civil society and empowerment of the local population.

(For detailed descriptions of concepts as local governance and dimensions of decentrali-

sation, see Land and Hauck, 2003). Administrative decentralisation and improved local

governance is considered important for poverty alleviation, based on the assumption

that it leads to better planning and execution of interventions, takes account of local

priorities, involves local responsibility and people’s participation.

Consequently, poverty and poverty alleviation has been analysed and assessed as to their

economic, social and political dimensions. Hence, the contribution to poverty reduction

was the key criterion in assessing aid effectiveness.

2.3 Scope of the study

The study was conducted in two districts in Tanzania that form part of the district pro-

gramme receiving Dutch aid. Political developments following Tanzania’s independence

in 1961 are taken as the point of departure, but the influence of the colonial period is also

taken into account.

Dutch support for district development dates from 1987, and an attempt is therefore

made to describe the poverty situation in the districts at three points in time: in the early

1960s when Tanzania became independent, in the mid-1980s when Tanzania formally

ended its socialist development strategy and a new policy for decentralisation and local

governance was announced, and at the present time after some twenty years of economic

reform.

9



Mbulu District in the Arusha Region and Songea District in the Ruvuma Region were

selected for the study. The choice of these districts was based on their peripheral location

and their poor communications with the country’s major economic centres, their internal

socio-economic differentiation, and the protracted period of the Dutch aid programme.

Considerable differences in population density between Mbulu and Songea may indicate

a great variability in the availability of land per capita. It has been assumed that these

would affect the nature of the poverty situation.

This evaluation is experimental in several respects: it attempts to reveal the nature of

long-term processes of change and their effects on poverty; it relates the relevance of the

aid programme to the people’s perception of poverty and its basic causes; and it consi-

ders the nature of that relationship as being crucial for the effectiveness of aid. Moreover,

the perception study by regular interviews was preceded in one of the two districts by

video recordings of people expressing their views on poverty, government policy, and

donor support. A compilation of these was subsequently shown to the people in some of

the villages concerned, who were invited to comment. In its approach, therefore, this

evaluation is close to what Booth labelled ‘a contextual method of gaining knowledge’,

i.e. an attempt to capture a social phenomenon, such as poverty, within its local social,

economic and cultural context (Voipio & Hoebink, 1999, pp. 37-38).

Due to its emphasis on perception and to the problems involved in analysing long-term

processes of change, the study has several limitations. Firstly, the actual poverty situation

has been recorded and analysed at district level rather than at the level of the selected

villages in the two districts. This has the advantage of linking-up the poverty situation

with the District Rural Development Programme, but prevents comparison of respon-

dents’ perceptions of poverty to their actual resource position, income and living condi-

tions. Second, long-term income trends could not be traced at the district level. Moreover,

information on social services was not readily available in the pertinent district offices in

Songea and Mbulu and could only partly be constructed because available statistics were

often inconsistent. This narrowed the analysis of poverty trends in the two districts.

Third, the division of Mbulu district into Mbulu and Karatu in 1998 hampered the use of

available statistics. The perception study was limited to villages in post-1998 Mbulu

District because of the expected high degree of poverty. Karatu is economically more

advanced and more dynamic; it has a much higher average income but also wider income

disparities.



10

2.4 Methodology

The following methods were used in the evaluation study:

a. Literature study in the Netherlands and Tanzania into the history of Tanzania and more

specifically into developments in Mbulu and Songea Districts. Public sources as well

as unpublished documents that provided specific information regarding the selected

areas were consulted.

b. File study of activities financed by the Netherlands over a period of some 15 years.

This occurred at three locations: the archives of the Ministry of Foreign Affairs and of

the Dutch Development Organisation SNV in The Hague, the archives of the Royal

Netherlands Embassy in Dar es Salaam, and the offices of district councils and

district administration in Mbulu and Songea.

c. Video recordings. To record the perceptions of the various groups of the local population

regarding their own situation, approximately 200 persons were interviewed as indi-

viduals, as representatives of organisations or as a group in eleven villages in Mbulu

District, seven of which were selected by MDRDP and four were chosen at random.

The video team conducted structured and unstructured interviews separately for

men, women, youth, farmers and officials. The team also organised group

discussions with different categories of the population. Finally, they interviewed key

informants in district centres, such as politicians, businessmen, government and

DRDP staff and representatives of Non-Government Organisations (NGOs).

An 80-minutes summary was made of some 36 hours of video recordings and shown

to the villagers for verification purposes. The main subjects mentioned by the

population helped to define the questions for the field study.

d. A field study by a team of Tanzanian researchers was carried out in four villages in

Songea and four in Mbulu. These villages were chosen on the basis of diversification

in the agricultural production system and ethnic background on the one hand, and

the extent of people’s participation in the district development programme on the

other. The study teams conducted structured and unstructured interviews in the

selected villages, separately for men and women, farmers and officials. The teams

also organised focus group discussions with different categories of the population.

Finally, they interviewed key informants in district centres, such as politicians,

businessmen, government officials, staff of the District Rural Development

Programme (DRDP) and representatives of Non-Government Organisations.

11



This methodology implies that perceptions were obtained through both video recordings

and analysis and a field study applying various types of interview. Subsequently, the

perceptions of the population were compared with other sources of information, such as

statistics, and progress and evaluation reports. The results of the field study were

produced in an internal report titled: Location-Specific Poverty Alleviation in Songea and Mbulu

Districts: A Research Report, 2001.

The various sub-studies were co-ordinated by two Dutch consultants and supervised by

the responsible IOB staff.

2.5 Structure of the report

The report consists of the following parts:

• Main findings of the study and issues for the future planning of poverty alleviation

programmes (chapter 1) and the objectives and evaluation approach (chapter 2).

• A background section which provides a description of the donor community’s views

on poverty and development (chapter 3), changes in rural Tanzania since indepen-

dence (chapter 4), poverty and government policy in Tanzania (chapter 5), and Dutch

aid and poverty alleviation in Tanzania (chapter 6).

• The next section contains a description and analysis of socio-economic change, and

of poverty and poverty alleviation measures at the level of the two districts. Chapter 7

describes the socio-economic and administrative changes in Mbulu district, and the

perceptions of different segments of the population regarding poverty. Chapter 8

deals with the Dutch aid programme in Mbulu district and provides information

about objectives, approach, activities and results of efforts to support development in

relation to people’s perception of the main causes of poverty. Chapters 9 and 10 con-

tain similar information for Songea district.

• Based on the descriptions in the preceding chapters, the final chapter (chapter 11)

assesses Dutch support for district development and compares these programmes

with the priorities of the population and popular perceptions of the relevance of

donor support.



12

3 Poverty, development and aid

3.1 Poverty and development

At present, three dimensions to poverty are commonly distinguished: economic, social

and political. In economic terms, poverty refers to the effects of the unequal distribution

of resources, such as capital and land, and the inadequate exploitation of those

resources. The social dimension relates to inadequate access to amenities that determine

the living conditions of people, e.g. drinking water, education and health facilities.

Finally, the political dimension implies the involvement in decision-making on the part of

the poor. These three dimensions are inter-related and mutually reinforcing. The degree

of poverty is measured with the use of indicators that have been developed by donors and

by governments in developing countries. The latter justify interventions and financing of

activities in the area of poverty reduction on the basis of these indicators. However, indi-

cators and interventions do not necessarily correspond with the perceptions of the target

population for whom the support is intended.

Over the years, views in development theories regarding the causes of poverty have

changed significantly and, implicitly also the strategies towards poverty alleviation.

In the 1960s, poverty was seen as a situation of underdevelopment to be solved by

economic growth. Poverty alleviation was therefore pursued by promoting economic

growth in developing countries. It was widely assumed that the effects of growth would

trickle down to all segments of the population and that poverty would gradually dis-

appear. For this reason, the (average) per capita income of the population was taken as

the major criterion for the level of development and the incidence of poverty.

When, in the 1970s, it was shown that poverty continued to exist on a large scale, even in

countries with economic growth, emphasis in the debate shifted to the need for direct

alleviation of poverty. Since the trickle-down effect did not seem to have worked as

expected, poor groups should be identified and would be provided directly with the

necessary basic needs (Streeten et al., 1981). At the same time, improvement of the

employment situation was considered a crucial aspect of fulfilling basic needs in the

longer term and, therefore, a major instrument for poverty alleviation (ILO, 1976).

The perceived need to deal simultaneously with poverty and economic growth led to the

formulation of a World Bank strategy which did not abandon growth as the primary

13


Poverty, development and aid

objective, but raised concerns about the distribution of its benefits. It made the case for a

better distribution of the fruits of additional growth and for an effort to secure the basic

needs of the poor and investments in human development (World Bank, 1974).

In the early 1980s, UNDP developed a series of indicators with which to describe poverty in

its Human Development Reports (UNDP, 1996). At that time, however, less emphasis was

placed on alleviating poverty when many developing country economies got into serious

difficulties as a result of the combination of decreasing world market prices and high

indebtedness. Priority in development aid shifted to macro-economic problems (balan-

cing the government budgets, improving the balance of payments, reducing the role of

the state in favour of market forces). As economic reform programmes carried the risk of

being detrimental to poor segments of the population, they were supplemented with

specific measures to reduce such harmful effects and were labelled as adjustment with a

human face (Cornia et al., 1987).

The World Bank, which played an important role in providing financial support for the

economic reform programmes, continued to emphasise the significance of economic

growth as a necessary condition for reducing poverty. In the 1990 World Development

Report, the link between economic growth and poverty reduction was further explored.

Achieving a pattern of development that successfully reduces poverty requires, in the view

of the World Bank, policies that provide opportunities to the poor and enable them to

participate in growth. Policies must be attuned to three broad tasks:

(i) to encourage rural development and urban employment through, among other

things, lower taxation of agriculture, market liberalisation and public provision of

infrastructure;

(ii) increasing access of the poor to land, credit and public infrastructure and services,

and

(iii) a specific approach for resource-poor regions: policies that facilitate out-migration

to reduce population pressure and additional investments to meet basic needs and

maintain or increase yield levels, with due attention for the preservation of natural

resources.

Since then the poverty agenda has broadened further. Evidence of persistent gender

inequalities and of a higher incidence of poverty among women focused attention on the

gender dimension of poverty. Then, in 1998, the World Bank report, Assessing Aid, drew



14

attention to the relationship between poverty, growth and good governance and

institutions. In that study good governance refers to economic policies affecting growth:

inflation, the budget surplus and open trade. It is argued that aid is more effective in

countries with such policies (World Bank, 1998). Others have criticised this report and

concluded, on the basis of the same World Bank data, that such policies are likely to

reduce the growth effect of aid (Dalgaard & Hansen, 2000).

In addition to this shift towards good governance as an important pre-condition for

successful poverty reduction, other views emerged in the 1990s. Amartya Sen (1990)

considered development from the concept of human freedom (deprivation of basic

capabilities). Thus, poverty reduction would imply that lack of freedom of any nature

should be removed. This view shows strong similarities with that held by the Institute of

Development Studies, which advocates the development of different types of (capital)

assets (Carney, 1998), and with the livelihood strategy developed by DFID expressed in

access to natural, physical, social, infrastructural and political resources and services

(which is also an element of the World Bank approach to poverty reduction).

A striking development in recent years is that the economy is no longer considered as an

individual phenomenon with its own dynamics, but that economic practices are seen as

being embedded in a wider framework of social and political processes. According to the

World Development Report 2000-2001, Attacking Poverty, ‘opportunity’, ‘empowerment’

and ‘security’ are central notions to an understanding of the position of the poor.

Opportunity focuses on building-up assets by the poor and improving marketing

mechanisms to benefit them. Empowerment involves making state institutions pro-poor

and enhancing the role of social institutions in poverty reduction. Security deals with

mechanisms for mitigating the vulnerability of the poor.

At least two important changes with earlier frames of analysis have been noticed:

(i) a departure from a relatively mechanical point of view towards a more people-oriented

or micro point of view on poverty alleviation, and (ii) a shift towards acknowledging the

importance of social networks and the position of poor people in those networks. This has

been labelled as an institutional or social capital perspective on poverty alleviation and

development (Bastiaensen et al., 2002).

Apart from these conceptual changes, a new and more pragmatic instrument for poverty

reduction is the Poverty Reduction Strategy Paper (PRSP). A PRSP reflects a national

development strategy integrating goals of macro-economic stability, economic growth

15



and poverty reduction with quantifiable targets designed by national governments after

broad consultation with key social groups. PRSP was introduced by the World Bank and

IMF in 1999 as a device with which to direct resources made available through debt relief

programmes towards poverty-reduction expenditure. Bilateral donors see the PRSP often

as a necessary basis for achieving poverty reduction goals through sector support.

3.2 Poverty reduction and decentralisation

The general focus on good governance in development assistance led to renewed atten-

tion for decentralisation processes. The importance of decentralisation for poverty reduc-

tion is acknowledged in the World Development Report 2000-2001, which refers to several

advantages of decentralisation. Firstly, development programmes can be more effective

in meeting local needs if they can draw on local information. This can help in identifying

more effective ways of building infrastructure or providing public services, and increase

the efficiency of regulatory functions that are more easily performed at the local level.

Moreover, the increased flow of information from local communities to government staff

may increase government awareness of local needs. Second, decentralisation has a strong

potential for increasing popular participation, in planning and implementation of deve-

lopment activities. Third, it can increase accountability and transparency. Local monito-

ring can be very effective for ensuring that officials perform diligently, while decentralisa-

tion makes it easier for people to obtain information on budgets and the use of funds.

The report also mentions several caveats to decentralisation. “Most important is the

likelihood that, in settings with highly unequal local power structures, decentralisation

will only bolster the power of local elites. Also critical is the extent to which local

government agencies are able to work with communities in implementing decentralised

programmes. Decentralisation therefore has to be approached cautiously to ensure that it

benefits the poor” (World Bank, 2001).

A recent report for the OECD/DAC network on aid evaluation assesses donor support for

decentralisation and local governance, and concludes that support to decentralisation is

not necessarily an effective tool in poverty alleviation. The poverty focus in the evaluated

programmes was generally weak; poverty targeting varied considerably in programme

design and was entirely absent in some cases. Moreover, there was little or no evidence

that donor interventions contributed significantly to pro-poor outcomes in terms of

economic activity, service delivery, and social or regional equality. The absence of pro-

poor outcomes was often due to lack of commitment by recipient governments (Schou &

Steffensen, 2002). These observations are in accordance with the findings of a field study



16

in six countries: “..there is little evidence that democratic local government initiatives can

do much to reduce poverty. The main reason is when governance is decentralised, local

elites get most of the power and steer benefits to either themselves and their interest

groups, or at least maintain the existing distribution pattern” (Blair, 2000).

The ways in which decentralisation has contributed to poverty reduction in a concrete

policy and country context in Africa has been explored recently in three country studies:

Ethiopia, Guinea and Mozambique (Bossuyt & Gould, 2000). Each of these countries has

lived through a long period of centralised one-party state systems. They are therefore

fairly representative for other African countries, for Tanzania in particular. Under these

one-party state systems, local structures lacked political autonomy and were intended to

bring the population in line with political doctrine rather than to increase political, eco-

nomic and social choices for the poor. Also, after the introduction of economic reforms

and the multi-party state in the three countries, government policy did not constructively

link decentralisation to poverty reduction and local institutions were not given an explicit

mandate to combat poverty. According to the study, the strategy underlying decentralisa-

tion was one of incorporation, masterminded and implemented by ruling elites in order

to maintain their control over people and resources.

The study identified several problems facing local government in executing its economic

and social development tasks, including poverty reduction: an unclear division of tasks

between central and local government structures, political dominance of centrally nomi-

nated officials, weak human material and financial capacities, and marginal involvement

of non-state actors. Despite these obstacles, the cases indicated that citizens and local

governments are beginning to occupy spaces created by reducing the role of the state.

Donors appeared to lack coherent strategies towards decentralisation and in their sup-

port to decentralisation a link to poverty reduction was hard to find. According to Bossuyt

and Gould: ’’…donor agencies are funding poverty-oriented projects of decentralised

public authorities and are supporting the capacity of local communities and civil society

organisations to collaborate with the local governments. However, these interventions do

not amount to a comprehensive, multilevel strategy that could fully tap the potential of

decentralisation for more effective poverty reduction”.

Based on a comparative analysis of donor support for decentralisation in the three

countries, Bossuyt and Gould present the following general recommendations:

17



1. Decentralisation best serves poverty reduction if it combines strategies of political

empowerment, resource mobilisation and enhanced service delivery.

2. Donor programmes should be incorporated in statutory structures, even if this

reduces short-term efficiency. Donors should not move away from local government

support because central government shows insufficient commitment to decentralisa-

tion.

3. Donors should not limit their support to public administration but should strengthen

civil society organisations in order to facilitate local communities’ access to informa-

tion and downward accountability of officials.

4. Donor support should help to engineer new relations between local governments and

the local population by involving both parties in planning and implementation of

development activities.

5. Donor support to central government is essential for an effective linkage between

decentralisation and poverty reduction. Crucial pre-conditions for successful

decentralisation have to be fulfilled at the national level such as a coherent legal

framework, a functional division of tasks between different layers of government and

access of citizens to an effective judicial system.

6. As individual donors are not in a position to provide the required support on their

own, there is a need for functional co-ordination and division of tasks among donors.

Recent experiences with donor support to local governance in Uganda provide a good

example of support to central agencies in charge of decentralisation, and effective donor

co-ordination under recipient country leadership (IOB, 2003). These experiences in

Uganda also confirm the limited effectiveness of decentralisation for poverty reduction

due to weaknesses of local institutions in service delivery. Public agencies and officers are

held in low esteem by the village population and the multiplicity of local taxes associated

with fiscal decentralisation is inimical to the expansion of monetary opportunities in rural

areas (Ellis and Bahiigwa, 2003).

3.3 Poverty alleviation in Dutch development co-operation

Poverty alleviation is the main general objective of Dutch development co-operation. In

the mid-1970s it arose as an explicit development aim in response to the modernisation

theory, in which economic growth through industrialisation was emphasised. The

assumed trickle-down effect of economic growth did not visibly improve the living condi-

tions of the poor, and Dutch policy therefore focused on direct poverty reduction through

the satisfaction of basic needs. Aid was to be directed to the poorest countries, and within



18

these countries to the poorest groups in society (the “poorest of the poor”). In addition,

Dutch policy underlined the importance of developing countries’ own policies to tackle

poverty. This meant a concentration on countries that addressed their poverty problems

seriously. This policy reorientation also changed the geographical pattern of bilateral aid.

At the end of the 1970s, Africa became the major recipient of Dutch aid (IOV, 1994).

In the early 1980s, poverty alleviation no longer occupied central stage in Dutch

development co-operation due to the economic recession and to the debt crisis in many

developing countries. Following the international development discussion, economic

liberalisation and the reduction of state intervention in the economy were combined with

poverty alleviation. In spite of this two-pronged approach, Dutch bilateral aid focused

increasingly on macro-economic programme aid, especially in respect of the African

countries. In 1984, the two objectives of economic reform and poverty alleviation were

combined into the general objective of “structural relief of poverty”, from the point of

view that sustainable improvement of the living conditions of the poor was not possible

without strengthening the national economies.

Criticisms of economic reform programmes, especially in the Adjustment with a human face

study commissioned by UNICEF, pointed to the potentially adverse consequences of such

programmes for the poorest groups in developing countries (Cornia, 1987). This had its

impact on Dutch aid. Employment creation and the improvement of social services

returned to the agenda and aid for the social sectors, often in the form of project aid, was

expanded. At this time, i.e. the second half of the 1980s, renewed attention was paid to

direct poverty alleviation in the framework of area-based programmes with a rural

development orientation.

During the 1990s, two new aspects of poverty alleviation were emphasised in Dutch

development co-operation policy. Firstly, the perception of poverty was based on an

explicit mutual relationship between its economic, social and political dimensions.

Second, the micro approach to poverty alleviation was linked with the macro context in

which the poverty problem became manifest. This meant that, on the one hand, the

importance of a sound macro-economic policy was endorsed, while on the other, more

attention was given to local developments, the dynamics of local cultures and endoge-

nous processes, whereby people try themselves to break-out of the circle of poverty

(Ministry of Foreign Affairs, 1991).

19



This policy was closely linked with the international consensus on development co-opera-

tion pursued through several international conferences, which finally resulted in the

Millennium Development Goals. The Netherlands joined that international consensus by

supporting the debt relief programme (HIPC) related to the Poverty Reduction Strategy

Paper (PRSP), in which national government funds were designated for priority sectors in

which poverty reduction goals were explicitly addressed.

A set of guiding principles were adopted for Dutch poverty reduction policy, such as

(i) poverty is not a static situation but the result of a process; (ii) poverty is highly context-

related, its analysis must address the specific situation in the country concerned and be

made by those directly involved; (iii) there are wide disparities between types and levels of

poverty; (iv) poverty reduction measures must be applied in ways and in sectors where

they will have the greatest impact, and (v) governments of recipient countries must make

explicit pro-poor choices in their development policies.

In addition, a recent document re-emphasises that poverty reduction policy should not be

limited to assisting the poor directly, but must also be oriented towards the macro con-

text (micro-macro approach). As intervention at the macro level ultimately achieves more

for poverty reduction than do large and incoherent groups of projects, aid is shifted

towards sector support for recipient governments. A pre-condition for sector support is

that the recipient government adopts a national poverty reduction strategy which has

been worked out in consultation with civil society, has a high degree of ownership by both

government and population, and is based on a comprehensive analysis of poverty-related

processes (Ministry of Foreign Affairs, 2001).

In Dutch development co-operation poverty alleviation is also linked to decentralisation,

which is considered an instrument of good governance and an aspect of democratisation

that may contribute to poverty alleviation. For this reason, the Netherlands supports

decentralisation processes, although it acknowledges – in line with World Bank views –

that there is no direct link between decentralisation and poverty reduction. The effects of

decentralisation for poverty reduction may be hampered by a low degree of organisation

of the poor to influence policy, the negative attitude of civil servants and politicians

towards the participation of civil society in development, the lack of funding and staff

with which to improve public services and the people’s reluctance to pay taxes for service

improvement due to their lack of confidence in the local administration (Ministry of

Foreign Affairs, 2002).



20

3.4 Summary and conclusions

The reduction of poverty has been the principal objective of Dutch development co-opera-

tion for the past 30 years, although the strategy for its realisation has changed several

times during that period.

Over the years the World Bank has played a dominant role, both in providing financial

support for development strategies and in analysing development processes, including

the poverty issue, and the factors that influence those processes. Dutch development

co-operation has generally followed the global trends set by the World Bank. The focus

shifted from the direct poverty alleviation of target groups in poor countries (1970s) via

the structural relief of poverty through macro-economic reforms and the reduction of

state intervention (early 1980s), back to direct poverty alleviation through improved social

services (late 1980s) and, subsequently, to linking the micro or project approach to pover-

ty reduction within the national or macro context in which the problem became manifest.

This resulted in a sector-wide approach in development co-operation, and to the selection

of social sectors such as education and health as priority sectors in most priority coun-

tries for Dutch bilateral aid. Sector support is linked with a national poverty reduction

strategy adopted by the recipient or partner country in order to enhance the effectiveness

of aid.

In practice, poverty has been reduced chiefly in East and South Asia under conditions of

sometimes spectacular rates of economic growth. In contrast, Sub-Sahara Africa has

experienced persistent poverty during the past four decades. Bad governance, wars and

corruption have frequently been mentioned as the main causes of poverty in this

sub-continent.

The identification of good governance as a major factor that influences the development

process in general and the effectiveness of development assistance in particular, has

renewed interest among donors for the decentralisation of authority to sub-national

levels of administration. Decentralisation combined with the improvement of local

governance has been considered an effective instrument for poverty alleviation because of

a greater awareness of local needs, the potential for popular participation in planning

and implementation and the increased transparency and accountability of the local

administration towards the local population.

In practice, support to decentralisation has not been the effective tool for poverty alle-

viation that was originally envisaged, mainly due to the generally weak poverty focus in

21



decentralisation programmes and to a lack of commitment on the part of recipient

governments to address the urgent needs of poorer segments of the population.

In sum, the conceptualisation of poverty and the identification of its underlying causes

have received ample attention, as have the accurate measurement of poverty and the for-

mulation of policies and strategies. On the part of the recipient country, policy formula-

tion has emanated in Policy Reduction Strategy Papers, that are linked to donor support

through the Sector-wide Approach.



22

4 Changes in the economy and localgovernance in rural Tanzania

4.1 Introduction: rural Tanzania before Independence

The basis for the structure of present-day rural societies in Tanzania was laid in the 19th

century. Agricultural systems were developed with a combination of food crops for self-

support and cash crops for sale on the market, supplemented by cattle, that served as a

source of food and as savings. Food crops such as maize and rice, cooking bananas,

cassava, sweet potatoes and various types of fruit were added to the African staples of

millet and sorghum. Also in the 19th century, farmers started to use iron farm imple-

ments and applied several other agricultural innovations: animal manure to maintain soil

fertility, double cropping and crop rotation systems, and irrigation. These resulted in

higher yields and made it possible to build-up food reserves, which gave the farmers more

food security. Under the influence of growing production, chiefly limited by available

labour, and expanding trade, the size of political units increased.

From the late 19th century onwards, new commercial crops such as coffee, cotton,

tobacco, and later sisal, pyrethrum, wheat and sunflowers were introduced under the

influence of the colonial powers. The colonial state of what was then German East Africa

(Tanganyika) needed income, and the Germans therefore tried to set up a capitalist

plantation agriculture, but with little success. Only where cash crops were introduced on

a small scale and by independent farmers was this achieved. In several districts, cotton

was introduced as a compulsory cash crop in order to facilitate taxation of farmers.

Many authors associate the German time with the outbreak of epidemics, especially

rinderpest (which in five years time killed off 90% of the cattle in most areas), smallpox,

and famine. Famine, combined with the loss of human life that came with the German

colonisation, claimed some three quarters of a million victims in five years, followed by

other famines after the Maji Maji rebellion (1805-1906) and World War I, especially in the

south of the country. As a result the rural population was probably worse off in 1920 than

in 1850 (Hyden, 1980; Coulson, 1982). Under German rule (until 1917), also an extensive

infrastructure of railroads, roads, towns, schools, churches and hospitals was built.

23


Changes in the economy and local governance in rural Tanzania

After the German defeat in World War I, Tanganyika became subject to British rule as a

mandate of the League of Nations. The British focused on extension and other support to

“progressive” farmers on comparatively large farms, which in turn would set an example

for the remaining rural population. As a result of this policy, the production of maize, cof-

fee and cotton increased dramatically between 1924 and 1934 (Hyden, 1980). However,

mono-cropping often caused soil erosion, disease and vermin in the crops. Also in the

late 1930s the tsetse fly spread to almost two-thirds of the country as a result of depopula-

tion caused by epidemics and cattle disease. Policies introduced to curb tse-tse

infestation and soil erosion were based on coercion; they did not take account of local

knowledge and were ineffective.

It was not until the mid-1950s that the British tried a different approach, that of “persis-

tent persuasion”: i.e. the growing of export crops was stimulated by extension based on

agricultural research. In addition, the government promoted co-operatives in order to

improve produce marketing. Ultimately, all this resulted in a significant expansion of

commercial farming on small farms, especially in those parts of the country with a

favourable resource position. The considerable demand for tropical agricultural produce

on the world market in the 1950s resulted in price increases and thus raised incomes for

small farmers (Iliffe, 1979).

4.2 Socialism and rural development 1961-1980

Upon Tanzania’s independence in 1961, the government inherited a commercial, agricul-

tural sector with an average annual export growth of more than 7%, a similarly growing

non-monetary farming sector, and a growth in gross domestic product of 4.8% per

annum. African commercial farming was characterised by inequality, however, which the

government felt was undesirable. When, after independence, a great deal of settler-

agriculture disappeared or deteriorated, the new government initiated an agricultural

development policy with a two-pronged approach, inspired by the World Bank (1960) and

included in Tanzania’s first Five-Year Plan (1961-1966). The improvement approach

comprised better services for commercial farming to large numbers of (small) farmers:

co-operatives for input supply and marketing of produce, agricultural extension to

improve farming practices, credits and linking of agricultural growth to community

development. The transformation approach involved support for commercial agriculture

controlled by the government in settlement schemes, which went hand in hand with

attracting migrants to unpopulated or sparsely-populated high potential areas.



24

The transformation approach was a dismal failure and the improvement approach had no

visible impact on agricultural productivity. Although agricultural production continued

to rise, this was chiefly the result of area expansion coupled with population growth.

While the percentage of subsistence farming in GDP remained approximately constant in

the years 1960-1968, the same period saw an increase in the share of subsistence farming

in the agricultural sector as a whole (from 55 to 71%). Agriculture was thus de-commer-

cialised, partly due to lower prices on the world market, partly to taxation and partly to

the creaming-off of surpluses in the marketing of agricultural produce by a marketing

apparatus that was managed inefficiently by the government. The marketing of agricul-

tural produce became more expensive for the farmers then it had ever been during the

colonial period (Hyden, 1980).

After the Arusha Declaration in 1967, the state took control of the economy through the

nationalisation of crucial sectors and the introduction of various forms of communal

agriculture in rural areas. The idea behind the central concept (Ujamaa) was that mutual

help, as practised within the African family, would be repeated at the village, district and

national levels. The policy was to be translated into decentralisation of government,

participative self-rule at the village level, and the provision of basic facilities to all

sections of the population. The most drastic aspect of this policy was the establishment of

communal village production units, within the scope of what was called the ‘villagisation’

programme. In Ujamaa villages, farmers were to produce and live in a commune. This

concentrated living in villages would facilitate social services (schools, health centres,

water pumps). The villages were to become the core units of multifunctional co-opera-

tives, above which umbrella co-operative unions were set up in every district, which – like

the village co-operatives – had an important political function. The co-operatives

acquired a monopoly on the sale of agricultural produce and the supply of inputs. In

1976, they were replaced by semi-government organisations (parastatals) controlled from

above and with the same functions. Agricultural production was stimulated in the form of

campaigns, accompanied by regulations for the compulsory planting of crops. Hired

labour was restricted, inter-cropping was sometimes prohibited and weeding became

compulsory.

Beginning in the 1970s, Tanzania’s economic growth slowly declined: 5.5% in 1965-1969,

3.6% in 1970-1974 and 3.4% in 1975-1980. Agriculture’s share in total investment fell

from 23% in 1966 to 13% in 1973 and some 8% during the second half of the 1970s.

The principal causes of the decline in output were decreasing prices for export products

25



on the world market, repeated droughts, and the disrupting consequences of the villagi-

sation policy. Agricultural price policies favoured food crops over export crops, and

farmers sold their produce as much as possible through informal market channels.

The expansion and improvement of social services in the rural areas was forcefully

pursued, with priority for the new villages. Policies regarding the improvement of social

services were fairly successful in the initial stage, and important results were achieved in

the education and health care sectors. From the mid-1970s onwards, however, there were

major shortages of medical instruments and medicines, while the primary education

sector suffered from too few qualified teachers and teaching materials. Maintenance of

the physical infrastructure also left much to be desired (IOV, 1994).

4.3 Economic liberalisation 1980-2002

As of the end of the 1970s and in particular after the war with Uganda (1978-1980),

Tanzania’s economy deteriorated rapidly. Between 1980 and 1985, growth in GDP fell to

an average of 1.1% per year. Government spending rose faster than its income and the

foreign debt took on uncontrollable proportions. The consequences became especially

apparent in rural areas, where agricultural production underwent deep crisis. As a result

of import restrictions, inputs were scarce, the export of cash crops stagnated, and food

production lagged behind population growth. The weak performance of various economic

sectors since the mid-1970s, mirrored in the stagnation of exports and GDP growth, has

had important consequences for Tanzania’s balance of payment position and debt situa-

tion. In 1985 only 30% of imports were covered by export revenues. The increasing gap

between exports and imports has largely been financed by foreign aid (IOV, 1994).

After a period of crisis management through unplanned austerity between 1979 and 1982,

and a transition period between 1982 and 1985 during which Tanzania attempted to

combine economic adjustments and its socialist policy, the country drastically changed

its development strategy in 1986. The new strategy entailed a return to a market-oriented

economy, sharp devaluation of the overvalued currency, and reform of the system of

foreign exchange allocation. Measures included a general reduction of state intervention,

the rationalisation of state industries, and liberalisation of agricultural marketing with

higher proceeds for producers (Doriye & Wuyts, 1992).

Initially, economic growth increased after introduction of the economic reform policy

(ERP). After the mid-1980s, food production improved at a fast pace as a result of the free



26

market, higher market prices and favourable weather conditions. In the early 1990s, how-

ever, food production again decreased due to lower input use caused by the abolition of

subsidies and decreasing market prices, but also due to the absence of any major techno-

logical breakthrough to increase the productivity of land and labour. The average value of

the six major export crops showed strongly fluctuating yields during the period 1980-

1990, due to varying world market prices and the consistent dominance of government in

the marketing boards. A 1992 study indicated that economic liberalisation had positively

affected village economies, meaning diversification in income-generating activities (no

export products), price levels and local trade, and thus the availability of reasonably

priced household goods at the village level (Booth et al., 1993). In general, the

performance of the Tanzanian economy was disappointing in the first half of the 1990s,

however, as average growth lagged behind population growth.

From 1995 onwards, the implementation of reforms was again given serious attention:

inflation rates went down from 25% to 5% per annum and structural reforms in the public

sector enabled a much larger participation of the private sector. This had a positive effect

on economic growth, which reached 4% in 1997 and surpassed 5% in 2001. Although for-

eign investment increased – with positive effects on exports (especially gold and fish) and

on tourism - overall investments stayed below 20% of GDP. Agricultural growth remained

below the target of 5% per annum. And although overall per capita income rose modest-

ly, output growth in agriculture was insufficient to make an impact on rural poverty

reduction.

Since the 1970s, Tanzania has been famous for its achievements in the social sector, in

particular health and education. At the end of the 1970s it was claimed that three-

quarters of the population lived within 5 km of a nearest health facility, and that major

improvements in average life expectancy and a significant drop in child mortality rate had

been realised. Enrolment rates in general education amounted to 90%, and literacy rates

were claimed to be as high as 93% among the male and 88% among the female popula-

tion. The economic crisis of the 1980s had a serious effect on the health and education

systems. Financial constraints caused a lack of basic supplies and equipment of health

facilities, while the country faced a rapid spread of Aids. In education, enrolment rates

decreased and literacy rates declined. Government expenditure on education fell dramati-

cally (from 14% of the national budget in 1975 to 7% in 1992). The downturn in health

care was less dramatic, but here too expenditure decreased from 7% in 1975 to 6% in 1992

(IOV, 1994). Moreover, in both sectors expenditure was distributed unequally over the

27



country. For example, the bulk of expenditure on health care was spent in the towns on

hospitals and curative centres aimed at those with higher incomes (Bol, 1997).

This declining trend in social services provision and social welfare did not change sub-

stantially during the 1990s. Literacy rates did not increase noticeably, and access to and

the functioning of health facilities did not show significant improvement; only the supply

of safe drinking water was improved.

4.4 Decentralisation and local governance

In the 19th century, social change in rural Tanzania occurred primarily as a result of

external influence. The centralisation of political power was associated chiefly with the

growing slave and ivory trade between the interior and the coast. Most small agricultural

communities were under the authority of a village chief, who checked trade routes or

created defence mechanisms against armed intruders. Although the size of political units

increased under the influence of this growing 19th century trade, there were only a few

large ethnic units. Therefore, when the German colonial power occupied the country at

the end of the 19th century, there was little resistance. The Germans divided the country

into districts and their government officials were assisted by Tanzanian secondary school

graduates.

In 1926, the British administration introduced the Native Authorities Ordinance, based

on the Indirect Rule system, with the British District Commissioner as representative of

the national government and local traditional chiefs to exercise administrative and judi-

cial duties at the level of the chiefdom. The major function of the chiefs was to keep order,

collect taxes and set up social and infrastructural services. Under the chiefs were sub-

chiefs for sub-divisions of the chiefdom and a headman at the village level. The main

function of the latter was to maintain law and order, collect taxes and mobilise his people

for community services, such as road construction and maintenance. As from 1950, other

civilians could also be appointed by the District Commissioner and traditional chiefdoms

were accepted as legal organisations. In several parts of the country, district councils were

established with a gradually expanding membership of civilians other than chiefs and

headmen. In 1953, the position of the traditional chiefs was confirmed by the African

Chiefs Ordinance, while the Local Government Ordinance provided that non-Africans

could also be included in the council.



28

After Independence in 1961, the chiefs were replaced by councillors, the majority of whom

were popularly elected, while others were appointed by the Minister for Local

Government. This situation changed in 1965 with the introduction of the one-party

system. Henceforth, the chairman of the council was no longer elected; the position was

filled by the party chairman in the district. The party (initially the TANU and later the

Chama cha Mapinduzi or CCM) was represented down to the village level through Village

Development Committees that replaced the dissolved traditional authorities. A further

intensification of state control was introduced when village populations were grouped

into Ten-House Cells, each with its own chairman.

In 1982, a Local Government Act was passed and a system of local governance was gradu-

ally created, although still under the leadership of the CCM until 1992. The multi-party

system was then introduced and in 1998 the Policy Paper on Local Government Reform

was published, specifying the design and endorsement of the local government reform

agenda. Since the late 1990s, regional administration has been restructured and its role

redefined, emphasising its supportive, advisory and facilitating role to local authorities.

Moreover, a common basket fund has been established to facilitate donor contributions

for implementation of the reform programme and a financial management guide pre-

pared. In 1998 a separate Ministry of Regional Administration and Local Government was

created and a review of training needs and arrangements for local government staff was

made.

The 1982 Local Government Act was revised in 1999. The new legislation involves the

devolution of powers to elected bodies, including powers to levy local taxes and the

obligation of central government to supply local governments with adequate grants.

In addition, the Act clarifies the relationship between central and local government,

promotes openness of council decisions and makes it accountable to the population.

Moreover, local governments have greater independence regarding the use of financial

resources made available by national government. Finally, in 2000 the Local Government

Reform Programme (LGRP) started and a special team was recruited to assist the new

ministry in managing the reform process.

Main objective of LGRP is to improve the quality of and access to public services provided

through Local Government Authorities (LGA), with the ultimate aim to contribute to

poverty reduction. This includes (i) shifting responsibilities for managing and providing

services from central to local government authorities (at district council level) and (ii)

29



increasing efficiency by reorganising the local administration and allowing councils

greater freedom in organising their activities and managing their personnel.

A joint government-donor review of LGRP in 2001 concluded that decentralisation policy

was not implemented consistently, and that progress was slower than anticipated,

particularly with the inter-related reforms in the areas of fiscal decentralisation, human

resources and legal harmonisation. The review led to the creation of three task forces to

address the issues in these three areas. Moreover, the emphasis of LGRP changed to one

of pre-qualification, whereby the reform process is implemented in those districts that

meet the ‘access criteria’, i.e. minimum conditions in terms of organisational capacity to

qualify for greater autonomy.

Major activities undertaken in the areas of fiscal decentralisation and human resources

are: (i) design of a formula-based system for recurrent sector grants for national develop-

ment priority areas; (ii) design of a formula-based system for capital grants –mainly

aimed at investments for infrastructure – and related to this a system for capacity-buil-

ding grants; (iii) analysis of financial management performance of LGAs to identify those

that qualify to receive block grants and associated greater autonomy and (iv) design of an

institutional strengthening programme of Regional Secretariats. Not much attention has

yet been given to define criteria to appraise whether the granted autonomy results in

improved and more efficient service delivery.

It is foreseen that qualifying LGAs will be granted fiscal autonomy in the second half of

2004. Support for capacity building will be provided to those that have not met eligibility

criteria.

The present system of local governance is organised down to the sub-village level. Within

villages there are sub-villages led by a chairperson who is elected by all adult members in

the sub-village. In principle, the sub-village assembly meets monthly. There is also an

assembly at the village level, i.e. a meeting of all adults ordinarily resident in the village,

and an elected Village Council. This latter consist of all chairpersons of sub-villages and

other members elected by the village assembly, of whom no less than one quarter should

be women. The Village Executive Officer is secretary to the council and the assembly.

The village council also meets monthly.

The forum at the ward level is the Ward Development Committee (WDC) consisting of the

Ward Councillor as chairperson, all chairpersons of Village Councils in the ward, and any

person who is member of the district council and resident in the ward. The Ward Executive

Officer is secretary to the WDC.



30

The District Council constitutes the highest political authority, consisting of elected

members (one for each ward), Members of Parliament in the district and other appointed

women members. The District Executive Director (DED), a civil servant appointed by the

Ministry of Regional Administration and Local Government, is secretary to the council

and has the support of technical staff. The Council may delegate responsibilities to ward

committees and village councils

The District Council is responsible for the planning and implementation of development

activities in the district. At (sub-)village and ward level, the planning process is a joint

exercise of technical officers and elected representatives. At the district level, proposals

are forwarded by the various departments to the District Planning Office and discussed by

the Heads of Department meeting chaired by the District Executive Director.

Subsequently, draft proposals are presented to the council committee responsible for

finance and planning and later approved by the full Council (United Republic of

Tanzania, Dar es Salaam, 2001a).

Thus, under the revised law, decentralisation is a comprehensive process encompassing

political, financial and administrative dimensions, which allows for participatory devel-

opment planning. The process is still closely supervised by the CCM, which maintains its

powers in almost all districts. In numerous districts council members of the ruling party

have been elected unopposed, reflecting its strong position in Tanzania.


Since the end of the 19th century rural Tanzania has experienced a process of fundamental

socio-economic change. This has been characterised, among other things, by the intro-

duction of new crops, commercialisation of agriculture, exposure to external economic

forces due to the production of export crops, monetisation of the rural economy, and the

provision of social services such as education and health care. The intensity of the

changes has differed widely throughout the country, mostly in relation to the resource

position of the various areas and their network of communications. This spatial differen-

tiation in the process of rural change has caused considerable inequity in rural incomes.

Areas with less potential in natural resources and/or those that are inadequately opened-

up to other parts of the country have generally been characterised by a high degree of

poverty.

31



After Independence in the early 1960s, the Government of Tanzania initiated a rural devel-

opment policy in which agricultural commercialisation was to go hand-in-hand with wide

participation of the population in development. This policy led to intensive state interven-

tion in, and control over, production, input supply and marketing services. In addition, it

caused priority to be given to communal forms of production and heavy subsidies on

inputs, which were mostly supplied as grants by donors. It soon became clear that neither

of the policy goals would be achieved: production stagnated, the agricultural sector

de-commercialised, and rural incomes declined. During this period, social services were

expanded and greatly improved, bringing better living conditions to rural areas.

The economic decline in the second half of the 1970s, however, meant that these

standards of social services could no longer be maintained.

Economic reforms introduced in the 1980s initially caused increased agricultural output,

particularly that of food crops for domestic consumption. However, low world market

prices for export crops, the absence of any major technological breakthrough to increase

the productivity of land and labour, and inefficient marketing systems, hampered further

agricultural output growth, with a dramatic effect on agricultural incomes and thus on

poverty. Although the provision of social services has improved, partly due to increased

development assistance, their sustainability depends on tax revenue based on income

growth. Economic reforms have not yet led to the envisaged process of rural socio-

economic change.

The post-Independence concentration of powers in the state put an end to the gradual

expansion of local governance that had started in the mid-1950s. Decentralisation was

taken up again in 1982, although to a limited extent and under strict supervision of the

ruling party. After introduction of the multi-party state in 1992 and the publication of a

Policy Paper on Local Government Reform, a more substantial Local Government Reform

programme started. This included the formal devolution of powers to elected bodies,

including the authority to levy local taxes, the obligation of central government to supply

adequate grants for the appropriate functioning of local administration, and the creation

of a common basket fund to facilitate donor support for local governance at the sector

level. In practice, the process is still controlled by the dominant political party, which has

maintained its power in almost all districts.



32

5 Poverty and government policyin Tanzania

5.1 Poverty in Tanzania

The analysis of economic change in Tanzania during the 20th century provides some

information regarding the poverty situation in rural areas. It refers to improvements in

rural living conditions from the end of the 19th century onwards which were related to

technological innovations in agriculture and to the introduction of export crops. This

resulted in higher returns, greater food security and, in selected parts of the country, to

cash income from agriculture. Export crop production expanded during the first half of

the 20th century and in a few areas to large-scale commercial farming by African farmers.

The process was accompanied by the initiation and gradual expansion of community

services, education and health facilities in particular, partly through the activities of

missionary organisations, partly through government.

Since Independence, the per capita income of Tanzanians has increased marginally. Over

the period 1960-2000 real income, i.e. corrected for inflation, rose by 22%, i.e. less than

0.5% per annum. Between 1960 and 1976 there was an average income growth of 2% per

annum, but from then onwards a serious decline occurred and by the mid-1980s incomes

were back at the 1967 level. Despite some recent improvements, per capita income in

2000 averaged US$ 280. In other words the 1975 peak had not been regained (in real

terms) and no net income growth had occurred over the past 25 years. Moreover, the

average per capita income is less than US$ 1 per day, the indicator for absolute poverty.

Income levels in Tanzania have always been influenced heavily by world market prices for

major export crops, e.g. coffee, tea and cotton. The decline in the mid-1970s was caused

by the end of the coffee boom, but also by the dramatic failure of socialist experiments in

rural areas, villagisation and state intervention in agricultural marketing in particular.

Low world market prices and the partial implementation of economic reforms, especially

in the marketing of agricultural produce, explain the stagnation in the rural economy in

the early 1990s.

The trend in average incomes since Independence is shown in Fig. 5.1.

33


Poverty and government policy in Tanzania

Four household surveys provide a more detailed insight into poverty trends in Tanzania

over the period 1983-1995, i.e. the period of structural adjustment of the economy and

economic reform. These surveys suggest higher incomes than the official statistics for

GDP per capita, and indicate a growth of average household income and expenditure dur-

ing the decade. However, the declining incidence of poverty showed some stagnation in

the early 1990s. The surveys confirm that poverty is still largely a rural phenomenon: rural

households account for 92% of the poor. The most recent household survey (2000/01)

provides a good indication of what happened to poverty during the 1990s by comparing

the results with the outcome of the 1991/92 survey.

Table 5.1. shows results for rural and urban areas in the country, whereby the poor were

defined as people whose income was insufficient to satisfy minimum basic needs (close

to US$ 100 p.p.p.a. in 2000/01).

It is clear that only a modest decline in poverty occurred during the 1990s, largely caused

by a remarkable improvement of income and living conditions in Dar es Salaam. In rural

areas, where most people still live, poverty reduction was indeed very minor. The share of

rural people living in poverty remained the same, 87% in 1990 and 2000, with four



34

90

95

100

105

110

115

120

125

130

1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999

Index (1960=100)

Figure 5.1: Per Capita Income Growth (1960-2000)

Source: Bigsten et al.,2001

regions (Lindi, Mara, Singida and Shinyanga) being worst of. Poverty levels are related

strongly to the education of the head of household: the less educated, the poorer.

Farmers are also poorer than other professionals or artisans, as are women compared to

men. The poor also benefit less from social services than others.

Further insight into the poverty situation may be obtained from several social indicators

reflecting health conditions, literacy, and access to education facilities and to drinking

water supplies. Table 5.2 shows long-term trends for some social indicators.

The most recent household budget survey includes further details of social change and

access to community services during the 1990s (Table 5.3).

It appears that Tanzania’s early record in social development was quite impressive. Until

1980 all major education and health indicators improved noticeably from the 1960 levels.

35



Table 5.1 Poverty changes during the 1990s

Total Rural Dar es Salaam Other Urban

The poor (as % of population):

1991/92 39 41 28 29

2000/01 36 39 18 26

Source: Household Budget Survey,2000/01

Table 5.2 Trends in Social Indicators during the period 1960-2000

1960 1970 1980 1990 2000

Adult Literacy Rate (% population over 15 years) 10 33 79 70 71

Net Primary School Enrolment (% school-going

children of age groups) 25 35 53 48 59

Life Expectancy (at birth) (Years) 42 46 52 54 50

Infant Mortality Rate (per 1000 live birth) 152 129 101 105 100

Access to Safe Drinking Water (% population) 25 35 53 46 55

Source: Various Human Development Reports (UNDP) and World Development Reports (World Bank)

During the 1980s, however, stagnation began, causing a decline in educational achieve-

ments, a no longer improving infant mortality rate, and reduced access to safe water.

Stagnation continued during the 1990s, except for a very recent primary school enrolment

increase and some improved access to (safe) drinking water. Life expectancy has even

declined as a result of the devastating HIV/AIDS crisis, thereby returning to the pre-1980

level. Enrolment in secondary schools remained at a very low level (5%) throughout the

period, worse than in other African countries, and is only now increasing gradually as a

result of private schools.

The combined effect of the various indicators shows a dismal poverty reduction record.

Improvements across the board occurred until 1980, but then incomes started to decline,

later followed by stagnation and a decline in social services. Only recently (since 1995)

have some improvements again become visible, be it very gradually. And what applies to

incomes is also relevant to social indicators: the poor always score worse than the rest of

the population in terms of access to, and use of, social services.

In Tanzania, two development agencies have conducted surveys of how the poor think

about poverty (so-called participatory poverty assessments). The report Voices of the Poor,

involving 6000 people in 87 villages, revealed differences in perceptions of poverty among

different categories of the population. It is striking that the poor do not define their main



36

Table 5.3 Changes in Social Indicators during the 1990s

1991/92 2000/01

Adult men with any education (%) 83 83

Adult women with any education (%) 68 67

Adults literate (%) 70 71

Primary net enrolment ratio 53 59

Children aged 7-13 years going to school (%) 57 61

Secondary net enrolment ratio 5 5

Households within 2 km of a primary school (%) 66 63

Households within 6 km of a primary health unit (%) 75 75

Households with a protected water source (%) 46 55

Households within 1 km of drinking water (%) 50 55

Source: Household Budget Survey 2000/01

problems in terms of (low) incomes. Much more important, in their opinion, is access to

productive resources such as land, agricultural inputs and production-oriented services.

According to the study, the poor give priority to secure land rights, the availability of agri-

cultural inputs, credit and useful technology, good transport, access to markets, and pos-

sibilities for saving. Voices of the Poor also demonstrates the important role of “social capi-

tal” (trust, unity, participation) in successful developments at the village level (World

Bank, 2000).

In a 1997 UNDP study about local human development, with field surveys in eight villages

in Shinyanga Region, the poor in rural Tanzania are characterised by their dependence on

subsistence agriculture. They live in the more remote parts of the country with low rainfall

and less fertile soil. Only cash crop producers, a mere 37% of rural households, seem to

have benefited from the post-1985 liberalisation of the economy. Poor households are

also characterised by a higher dependency rate and lower levels of education. While

female-headed households on average do not seem to be poorer than male-headed ones,

the UNDP study states that they perceive themselves generally as being poorer.

The people emphasised social insecurity, weak social services and gender inequality as

factors that contribute significantly to poverty. Cultural thresholds (e.g. unequal access

to land and capital at the household level), weak governance, lack of funds and poor

infrastructure were also mentioned as obstacles to development (UNDP, 1997).

The views of Tanzanians about poverty are related to the way of life of the groups in ques-

tion. Material and non-material dimensions both form part of their perceptions, as well

as temporary and permanent characteristics.

In another study (Luwaitama et al., 1995) two types of prevailing “discourses” regarding

poverty are distinguished: fatalist and optimist. The fatalist discourse involves an apparent

resignation to the predominant definition of the poor by non-poor, as those who lack

‘inbred’ characteristics enjoyed by the non-poor: clever, lucky and hardworking. In the

optimist discourse, there is resistance to being called poor, a notion that is only deemed

applicable to those who are lazy, stupid, extravagant or insane. In this connection, it is

interesting to note that many women feel that men who squander the family income on

alcohol cause poverty in the family. The poor tend to consider those people as poor whose

social circumstances are desperate (widows, orphans, the homeless and landless, those

who fail exams, give up and live as underdogs). The lack of children and husbands is also

considered a sign of poverty.

37



According to this study, women tend to be less fatalistic and more optimistic than men,

and rural residents also seem less fatalistic than townspeople. Obviously, optimistic

people seem more capable of fighting for poverty reduction than the fatalists. To the poor,

the only ones that are poor are those who have given up the struggle for better, especially

materially better, living conditions (Luwaitama, 1995).

5.2 Poverty and government policy 1961-2002

After Independence, the Government of Tanzania put a high priority on poverty allevia-

tion. Together with ignorance and disease, poverty was considered to be one of the three

‘’enemies of development’’. In line with the view current in the 1960/70s about the role of

the state in development, state interventions were initiated to reduce prevailing economic

and social inequalities. Government policy therefore focused on establishing state

enterprises in the agricultural and industrial sectors, and on the construction and

improvement of health facilities, schools and water supply systems. In remote rural areas,

the dispersed population was concentrated into Ujamaa villages in order to facilitate the

provision of such services.

State control over the economy did not bring about the anticipated increases in produc-

tion, however. Decreasing world market prices and poorly functioning marketing services

reduced farmers’ incomes and resulted in a downturn of the rural economy. In turn, this

endangered the maintenance of community services; the ultimate effect was a decline of

those services, a deterioration of the living conditions of the population, and an increase

in rural poverty.

During the period 1980-1995 an explicit poverty reduction policy did not exist. The severe

economic crisis and the fall in development aid forced the Government of Tanzania to

reduce public expenditure and, after 1985, to initiate economic reforms. Policy then

focused on macro-economic issues such as an appropriate exchange rate, foreign

exchange earnings, macro-economic stability and the reduction of inflation, privatisation

and market liberalisation. Some attention was also given to the rehabilitation and

maintenance of basic social services. In general, therefore, government was preoccupied

with crisis management and economic reforms rather than with a comprehensive poverty

reduction policy.

In its economic reform policy, the Tanzanian government deployed three instruments to

redistribute income: taxes, trade regulation and credit facilities. The results were not

always favourable to the rural population. A significant portion of collected taxes was



38

used for debt repayment, recurrent expenditure in the social sector (salaries of teachers

and health personnel), and for keeping the destitute industry on its feet. At the same

time, government expenditure on the agricultural sector decreased, which must have had

significant consequences for productivity and poverty in the rural areas. Trade liberalisa-

tion positively affected the income from export crops and credit arrangements increasing-

ly became available for the private sector. As this involved only a small number of farmers,

however, it did not have any positive effect on rural poverty. Moreover, the effectiveness of

several safety net programmes for poverty reduction was restricted due to official

minimum wage regulations and the lack of any explicit targeting of the poorest groups in

society (Bol, 1997; Voipio & Hoebink, 1999).

In 1995 the Government of Tanzania established a task force consisting of experts from

government and civil society to design a long term vision for the country’s development.

The group designed Vision 2025, an ambitious strategy with long-term macro economic

and sector goals, including an economic growth rate of 8% per annum. With this target

in mind, a National Poverty Eradication Strategy was designed which aimed at halving

absolute poverty by 2010 and eradicating abject poverty by 2025.

In an effort to strengthen ownership of the development process, a Tanzania Assistance

Strategy was formulated to co-ordinate national and international (donor) efforts towards

poverty eradication. Moreover, as a result of a new Highly Indebted Poor Countries (HIPC)

initiative for debt reduction, a Poverty Reduction Strategy Paper was formulated in

1999/2000. This means that only since very recently, Tanzania has had a comprehensive

national strategy for poverty alleviation. (URT, 2000)

The government’s current policy concerning poverty alleviation is outlined in this paper.

It concentrates on efforts aimed at (a) reducing income poverty, (b) improving human

capabilities, survival and social well-being; and (c) alleviating the extreme vulnerability of

the poor.

Reducing income poverty will be pursued through sound macro-economic policies and by

intensifying the implementation of reforms for increased marketing efficiency, particular-

ly in agriculture, reforms aimed at promoting export-oriented expansion and diversifica-

tion, and efforts to raise investment (from 15% to 17% of GDP). The improvement of

human capabilities, survival and social well-being will focus on the rehabilitation of exis-

ting structures and raising the quality of services on offer. A crucial factor is the abolition

of primary school fees to ensure that children from poor families in particular will have

access to basic education. Local communities are expected to play a major role in identi-

39



fying the needs of vulnerable groups, which will then be addressed by existing govern-

ment programmes and by enlisting the involvement of donors (United Republic of

Tanzania, 2001a). The Tanzanian Government intends to restrict its financial intervention

to the following priority areas: (basic) education, primary health care, rural roads,

agriculture, (drinking) water, the legal system and combating HIV/AIDS.

In line with the PRSP, public expenditure in priority sectors expanded from 20% of total

expenditure in 1998/99 to over 30% in 2000/01 and to almost 45% in the 2001/02 budget.

To a large extent, however, this only rectifies a downward trend in the recent past: for

2002 social sector spending will only be 4% of GDP, whereas it was 6% during the early

1980s and also 4% during the early 1990s. Within priority sectors, some shifts towards

pro-poor items like primary education, primary health care and rural roads are also

planned.

Table 5.4 below gives some details about public expenditure in health and education,

priority sectors of PRSP and concentration sectors for Dutch assistance to Tanzania.

As poverty is largely a rural phenomenon, the Tanzanian Government has designed a

special Rural Development Strategy (December 2001), the overall objective of which is to

provide a framework for the implementation of sector policies concerned with the deve-

lopment of rural communities. It is stressed that rural development is largely dependent

on the growth rate of agricultural production. The increase of rural incomes through

improving the productivity of smallholders, pastoralists and non-farm enterprises is

therefore at the centre of the strategy, which includes intensifying and commercialising



40

Table 5.4 Government expenditure in the social sectors (TSH bn/1995 prices)

90/91 91/92 92/93 93/94 94/95 95/96 96/97 97/98 98/99 99/00

Recurrent Budget 515 542 481 578 521 472 505 562 560 581

Of which:

- Education 39 48 45 56 52 51 36 66 95 107

- Health 29 35 28 36 49 42 19 25 49 51

Sources: W.B., Country Economic Memo, 2001,Vol. 2

smallholder agriculture. The provision of basic infrastructure and services is considered

both as essential for rural communities and as a pre-condition for rural economic growth.

It includes access to assets such as education, health, land, financial services and mar-

kets, and ensuring sustainable management of the natural resource base. The strategy

has been worked out in an action plan and is complemented by an Agricultural Sector

Development Strategy (ASDS) (URT, 2001b).

The primary objective of the agricultural strategy is to create an environment conducive to

increasing agricultural productivity, to improve farm incomes and reduce rural poverty,

and to attract private investment. Interventions are focused on four key areas: (i)

strengthening the institutional framework for agricultural development, (ii) creating a

favourable climate for commercial, agriculture related, activities, (iii) improving support

services by the public and private sector, and (iv) strengthening market efficiency for agri-

cultural inputs and outputs. Implementation of the strategy will be pursued through

District Agricultural Development Plans (DADP). ASDS remarks, however, that this is

hampered by the non-existence of DADPs for most districts, and that agricultural

planning at the district level is an underdeveloped area (ETC, 2001).

Tanzania is highly dependent on donors for the implementation of the rural and agri-

cultural development strategies. An inventory has thus been made of on-going and

pipeline projects in the agricultural sector in order to establish their consistency with

ASDS. Area-based programmes, such as DRDP, are included in this inventory insofar as

support is provided to agriculture and livestock development. Usually, such programmes

represent a combination of institutional strengthening, support of basic services and

group formation, and participatory planning.

It appears from the inventory that the World Bank and IFAD are by far the largest funding

agencies, together covering more than half of total commitments. The Netherlands is

included among major donors. The majority of donor projects deal with the delivery of

support services and most have a component of institutional strengthening. Many pro-

jects focus on production rather than on productivity and profitability. The inventory does

not make any specific references to the strategy’s role in, and potential for, poverty reduc-

tion, but the rural development strategy sees agricultural growth as the crux of rural

development and rural poverty alleviation (ETC, 2001).

In 2003, the Advisory Council on International Affairs (AIV), an advisory body for the

Dutch government and parliament, made an analysis of the poverty reduction strategies

41



in three priority countries for Dutch aid in Sub-Saharan Africa: Ethiopia, Tanzania and

Uganda. For Tanzania the AIV report observed that the country’s policy with regard to

poverty reduction is unclear and that only a small proportion of public expenditure is

reaching the poor. Moreover, strategies for promoting development are said to be mis-

sing, while also more attention is needed for the productive sector, especially agriculture.

The report concludes that Tanzania’s PRSP is characterised by a focus on the social

dimension of poverty, remarkably little attention for the economic dimension, and a

narrow view on the political dimension, which is limited to the need for institutional

change rather than the social and political emancipation of the poor (AIV, 2003).


Tanzania is characterised by persistent poverty. Measured by changes in income and the

provision of social services, there have been only marginal improvements in the poverty

situation over the past 40 years. Since Independence, average per capita real income has

hardly increased, i.e. less than 0.5% per annum, and present real incomes are below

those of 1975. There are two main reasons for this: low world market prices for the coun-

try’s main export crops and ineffective government policies. The lack of agricultural diver-

sification meant that low world market prices continued to play an important role and the

vulnerability of the rural economy has not been reduced. Social indicators show a similar

trend: an improvement of the situation during the 1960s and 1970s and a subsequent

decline. Since the mid-1990s an upward trend in the distribution and quality of commu-

nity services has again become visible.

There is a definite gap between the indicators used for poverty measurement by govern-

ment and the donor community, and the perception of the population. The poor do not

define the basic problems underlying poverty in terms of income. Instead, they empha-

sise the importance of access to, and control over, productive resources: land and land

rights, inputs and production-oriented services, and the social insecurity that results

from their absence. In other words, the poor emphasise the underlying socio-political

factors of rural poverty. In their perception, conditions are unfavourable for earning an

income and services provided by government and the private sector are highly

inadequate.

The crucial factor in the combating of rural poverty, at least in the short term, is enhan-

cing the effectiveness of government policies. With the advantage of hindsight, it must be

concluded that Tanzania’s policies towards poverty alleviation have been quite ineffective.



42

Up to the mid-1980s policies were characterised by heavy state intervention in economic

and social life and a strong focus on the expansion of freely-provided community services.

As state control over the economy did not bring the expected benefits in economic growth

and income increases, and donors discontinued their support for Tanzania’s development

model, however, social services could no longer be maintained and living conditions

therefore started to deteriorate.

Subsequent economic reform policies lacked any explicit poverty focus, as did develop-

ment aid strategies, and the presumed advantages of market liberalisation did not mate-

rialise for large segments of the population, particularly in the rural areas. In other words,

government’s economic reform policies did not create an enabling environment for

agricultural production growth that could lead to reducing rural poverty. Only since the

mid-1990s, encouraged by donor incentives for debt reduction, has Tanzania formulated

a comprehensive poverty reduction strategy. This has been worked out in detail for rural

areas and the agricultural sector in separate policy papers; implementation has hardly

started, however, and investments necessary to execute the strategy depend almost

entirely on donor funding. Moreover, the political dimension of the poverty reduction

strategy lacks attention for the social and political emancipation of the poor. It remains to

be seen, therefore, how Tanzania’s good intentions will materialise in practise and what

the effects will be on the poverty situation.

The elaboration of a reform policy for rural areas is in itself an important step. Poverty in

Tanzania is chiefly a rural phenomenon and poverty levels are strongly related to the edu-

cation of the head of household. In addition, women are poorer than men and the poor

benefit little from social services. Geographically the poorest groups are concentrated in

remote parts of the country with low and erratic rainfall and less fertile soils, and are

dependent largely on subsistence agriculture. In such areas the potential for achieving

higher incomes through raising productivity and profitability in agriculture is therefore

very limited.

Nevertheless, these are the principal objectives of the agricultural strategy. Key interven-

tions mentioned in the strategy include improving agricultural support services and

strengthening marketing efficiency, corresponding with the priorities perceived by the

population. To achieve higher productivity, implementation of the strategy requires area-

specific interventions, which take account of resource potential and the poor’s access to

productive resources, and technological innovations. The strategy underlines the crucial

role of district agricultural plans in achieving rural development objectives, but acknow-

43



ledges that their design is seriously flawed by institutional weaknesses at the district

level. This implies that donor support to organisational strengthening at the district level

may help in the effective implementation of government’s rural development policy.

This can only be feasible, however, if public investments lead to a production environ-

ment that encourages local producers to introduce technological innovations that will

increase their land and labour productivity.



44

6 Dutch aid and poverty alleviationin Tanzania

6.1 Background

Since becoming independent in 1961, Tanzania has been the recipient of substantial

volumes of foreign aid, estimated at almost US$ 13 billion over the period 1970-1992 and

accounting for over 80% of net official inflow of external capital. From 1993 to 2000, offi-

cial foreign aid averaged US$ 1,016 million a year, giving a total of about US$ 21 billion

since 1970. The terms of the aid given to Tanzania have been quite favourable: about 75%

of the aid flow consisted of grants; the grant component increased from 50% in 1970 to

over 95% in 1990, decreasing after 1993 to almost 70%. Including the grant element in

loans, the country received over US$ 18 billion in grants between 1970 and 2000. Bilateral

aid dominated the package throughout the period: close to 90% around 1980 with a gra-

dual decline towards 70% thereafter. Aid fluctuated over the years in accordance with the

confidence that donors had in Tanzania’s development policies, as indicated by Fig. 6.1

The significance of foreign aid for the Tanzanian economy is thus enormous. Up to the

early 1990s two-thirds of imports were financed with external aid, 80% of the develop-

ment budget, and almost 40% of the current government budget. Since 1990 Official

Development Assistance (ODA) has fallen sharply, from 30% of GNP to 12.5% in 1998.

ODA represented US$ 46 per capita in 1990 and US$ 31 per capita in 1998; and around the

year 2000 total aid dropped to US$ 500m per year (Bigsten et al., 2001).

The sector orientation of aid changed over time. In the 1960s and early 1970s, most pro-

ject assistance was directed towards the agricultural and transport sectors. During the

second half of the 1970s, the focus of development assistance to Tanzania shifted from

agriculture towards industry and energy, although the Nordic countries in particular

continued to support the social sectors and rural development. Figures for the late 1980s

and early 1990s indicate transport and industry as being the main sectors for foreign aid,

together receiving almost half the total aid flow (IOV, 1994).

More important than sector-wise changes is the shift from project aid to programme

assistance, especially after the introduction of the Economic Reform Programme in the

45


Dutch aid and poverty alleviation in Tanzania

early 1980s. Programme assistance comprised commodity import support to make better

use of existing production capacity, particularly in the industrial sector, and general

macro-economic support that was largely channelled through the World Bank.

Programme aid was primarily aimed at restoring economic growth and, therefore, only

indirectly at alleviating poverty. In sum, it is striking that only a limited share of total aid

was allocated to direct poverty alleviation in rural areas, including education, health and

drinking water supply. Because of severe criticism of the WB’s emphasis on the economic

character of Structural Assistance Programmes, special programmes were designed to

reduce the disadvantages of structural adjustment for the poor.

To what extent did the massive foreign aid help to alleviate poverty? A comparative study

of several Tanzanian programme evaluations by various donors, undertaken during the

early 1990s, shows that project aid and macro-economic support did not bring the



46

Figure 6.1: Aid to Tanzania 1970-1996

Source: Bigsten et al., 2001

0

200

400

600

800

1000

1200

1400

1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996

Total (USD million)

0

20

40

60

80

100

120

140

Aid/Capita (USD)

Aid/capita

Total aid

intended results in terms of economic growth, improved income distribution and poverty

alleviation. With regard to project aid, it is noted that projects suffered from hasty prepa-

ration, were too large and too complex, were beyond Tanzanian managerial capability,

lacked adequate manpower, and were insufficiently supported by the Government of

Tanzania. Therefore, they had a low economic rate of return, and a high failure rate.

Macro-economic support or programme aid, although less donor-driven, was geared

towards a relatively small number of private firms and (semi-)state enterprises, chiefly in

the industrial sector and located in the main towns, especially in the capital. This aid had

little effect on rural areas, where most of the poor are living (Bol, 1995).

A more recent study that focused on the approach to, and effectiveness of, European pro-

ject aid for poverty reduction in Tanzania reaches conclusions more directly related to

poverty reduction:

(i) available statistics seem to contradict that donors have contributed a substantial

volume of aid to poverty reduction projects, including those in rural development,

education and health;

(ii) project objectives were mostly expressed in terms of income increases, improvement

of well-being or a move to sustainable livelihood, rather than in terms of poverty

reduction. Also most projects and programmes seemed to have little idea about their

impact on poverty reduction;

(iii) the larger percentage of donor supported programmes oriented towards

income–generating activities for poverty reduction are the multi-sector integrated

rural development programmes or district rural development programmes (Voipio &

Hoebink, 1999).

6.2 Poverty alleviation in Dutch assistance to Tanzania

Tanzania is by far the largest recipient of Dutch aid in Africa. During the 1960s and early

1970s the aid volume was fairly modest and provided on an ad hoc basis. In 1970 Tanzania

became a priority country for Dutch bilateral aid and the aid flow increased spectacularly

from an average of US$ 7 million (NLG 20 million) per annum to US$ 45 million

(NLG 100 million) in the second half of the 1970s. It further increased to US$ 65 million

(NLG 180 million) in the early 1980s, subsequently decreasing to US$ 45 million again in

the early 1990s. During the 1990s, the aid volume fluctuated between US $ 45 and

55 million (NLG 80 and 100 million). All in all, Dutch aid to Tanzania totalled no less than

US$ 1450 million, or NLG 3200 million, up to 2002.

47



Over time, the sector-wise distribution of Dutch aid has changed significantly. Between

1970 and 1990 the proportion made available to the manufacturing sector almost

doubled from about one-third to over 60% of bilateral aid. This change was closely

related to the increase of macro-economic support. The share of the agricultural and

livestock sectors was reduced from 20 to 10%. Over the period almost one-quarter of all

aid continued to be channelled to the social sectors (education, health, water and

community development).

The shift from project to programme aid during the period 1970-1990 and from agricul-

ture to industry also caused a geographical bias towards urban centres. It was estimated

that almost two-thirds of all Dutch aid went to urban areas, the capital Dar es Salaam in

particular. Only one-third was focused directly on the development of rural regions where

the vast majority of the country’s population lives (IOV, 1994).

The effectiveness of Dutch bilateral aid for poverty reduction was assessed in the IOV

country programme evaluation of 1994. The report concluded that:

During the 1970s and 1980s Dutch development aid to Tanzania was only to a minor extent

orientated towards direct poverty alleviation and the basic needs of the population. Less than

20% of all bilateral aid was focused on this main objective of Netherlands development assis-

tance. It was mainly pursued through rural development projects. These projects were of two types:

those promoting economic opportunities for the poor and those delivering social services (educa-

tion, health and drinking water) to the poor. The results of the production-oriented projects

primarily benefited those households that had adequate productive resources at their disposal.

In some programmes, special measures were taken to enhance the opportunities for the poorer

segments of the population. The severe economic crisis of the 1980s seriously affected the level of

social services: facilities could not be maintained and basic supplies and equipment were lacking.

The drinking water programme clearly shows the dilemma between improving social services for

the poor and the cost recovery principle. In sum, Netherlands development co-operation with

Tanzania contributed little towards direct poverty alleviation, partly due to the meagre budget

earmarked for it and partly owing to the low effectiveness of poverty-oriented programmes (IOV,

1994).

Since the evaluation of the country programme in 1994, Dutch development assistance

gradually became concentrated on a limited number of sectors and regions. Support to

the cotton and sugar sectors and assistance to state industries was terminated. Macro-

economic support continued to be important during the 1990s but was accompanied by a



48

gradual expansion of aid to district development programmes with a focus on rural

development. In 2000, there was a more drastic change in the aid programme following

adoption of the sector approach in Dutch development co-operation. After consultation

with the Government of Tanzania, the Netherlands decided to concentrate its support on

four sectors: Education, Local Governance, Health, and the Private Sector. The poverty

focus formed a major criterion in the selection of these sectors. The district development

programme was included as part of the support to Local Governance, and it was assumed

that emphasis on decentralisation and participation of the local population in decision-

making, as well as the delivery of services, would contribute to poverty reduction.

After lengthy discussions with experts and the pertinent ministry, agriculture was

excluded from sector support against the preference of Tanzania. Main arguments of the

Netherlands for not selecting agriculture were the deficiencies in the institutional and

policy framework, while it also felt that the agricultural sector would be better served by

incentives through local governance and the private sector (RNE, 2000, Annex 4).

In 2003 some further changes in the Dutch aid programme for Tanzania are emerging: a

decrease in the volume of bilateral aid in the order of 30%, a reduction of the number of

sectors supported and, as a consequence, some uncertainty about the magnitude and

nature of Dutch support to local governance.

The selection of the social sectors of education and health and the continuation of the

district development programmes under the heading of local government support

indicates greater orientation towards direct poverty alleviation after 2000. Consecutive

annual reports of the Netherlands Embassy since the late 1990s emphasise the aid

programme’s orientation and the range of inputs towards poverty reduction, but do not

contain any information on concrete results. Instead, the annual report for 2002/03 states

that “it is considered of utmost importance to refocus RNE’s interventions around three

strategic areas (good governance, institutional reform and pro-poor and equitable

development policy) within the general context of poverty reduction.’’ Recent annual

reports also mention that the effects of Dutch support for poverty alleviation should

primarily materialise through improved Tanzanian development policies. In other words,

present Dutch aid policy focuses on contributing to an enabling environment for poverty

reduction. Therefore, the relevant question in the framework of this evaluation is whether

or not the district development programmes indeed helped to alleviate poverty.

49



6.3 District programmes as a tool for poverty alleviation

District programmes in Tanzania are a sequel to the earlier integrated rural development

projects (IRDPs, in Tanzania known under the name of RIDEPs), which were highly

popular among donors in the early 1970s. IRDPs were a specific type of area-based

development projects, with the following main characteristics:

• the multi-sectoral nature of activities supported, usually implemented by several

agencies;

• the explicit reference to poverty alleviation and to the rural poor as the main target

group;

• the emphasis on local participation and mobilisation of the target group;

• the limitation of the geographic scope of intervention primarily to administrative

areas, districts in particular.

In general, such area-based development projects were considered a failure. Several

evaluation studies reached similar conclusions with regard to low effectiveness and the

lack of sustainable results. Results were particularly disappointing in Sub-Sahara Africa.

By the mid-1980s, they were held ‘in extremely low repute’ by many donors (Conyers et

al., 1988).

Most IRDPs in which the Netherlands was involved as a donor sought to accelerate the

development of backward and resource-poor regions. Weak institutional capabilities on

the one hand, and complex and ambitious development assistance plans on the other,

meant that IRDPs often showed a relatively high donor influence in their design and

implementation. Most were given assistance for a long period, i.e. for more than 10-15

years. Despite this long-term support, the assistance was often characterised by a strong

fragmentation of activities, short commitment periods and frequent extensions, and

disappointing results (Sterkenburg and van der Wiel, 1999).

The district development programmes, which started in the mid-1980s, show many

similarities to the earlier IRDPs: in poverty orientation, in geographical scope, in focus on

people’s participation, and in the multiplicity of activities supported. They are much

more inter-linked with decentralisation, however, and with attempts to strengthen insti-

tutions at the lower echelons of government administration. Decentralisation includes

the participation of the private sector and the strengthening of civil society. These new

area development programmes, therefore, have shifted the focus towards a more wide-

ranging development approach which takes account of all actors in the development



50

process rather than concentrates on government alone (Sterkenburg & Van der Wiel,

1999).

The reasons for the Netherlands to become involved in district development programmes

in Tanzania in the late 1980s were threefold:

(i) to provide more support for rural development;

(ii) to support the newly (re-)created Local Government Authorities in the hope that aid

could be brought closer to the people through these institutions rather than through

central government;

(iii) to concentrate Dutch aid in particular areas, preferably where it already supported

sector projects (Bol, 1999).

The overall aim of District Rural Development Programmes was to increase the income of

farm households and to improve the living conditions of the rural population. At a later

stage, this general aim became “to improve the well-being of the rural population in a

sustainable way”. This change in the overall aim reflects the shift of emphasis to social

services rather than to income-earning activities and related levels of living. More specific

objectives were formulated for the individual districts included in the programme and for

the various phases of implementation.

The programme started in 1987 in four districts: Mbulu, Bukoba, Maswa and Meatu, but

was gradually expanded to fourteen districts at present (see map 1). Total expenditure in

the period 1987-2001 increased gradually together with the incorporation of new districts

from US$ 0.6 million per annum in the 1980s to 3.7 million at the beginning of the 1990s,

reaching 11 million in 1998. From then on a fairly significant decline occurred to US$ 6.2

million per year in 2001. Financial assistance to DRDPs remained more or less constant

during 1997-2002 if expressed in Tshs. Over the period 1987-2000 expenditure on DRDP

equalled some 10% of total Dutch development assistance to Tanzania.

A recent review has observed that there were major variations in total funding among

districts. In addition, the gap between the budgeted and actual expenditure was quite

large but had improved markedly over the period. Finally, per capita expenditure varied

significantly from district to district, whereby districts with smaller populations appeared

to have higher per capita expenditure (RNE/SNV, 2000).

51



In strategy and approach, DRDPs were to be characterised by flexibility rather than by

blueprints. After a brief period of orientation, multi-sectoral development plans were

written by DRDP staff in co-operation with district planners, including financial

paragraphs for donor support. Starter activities were to be identified by the district



52

Map 1 DRDP supported by the Netherlands

administrations and focused on three broad categories: improvement of physical infra-

structure, increase of agricultural production, and strengthening of institutions.

The latter focused on government staff in the district, planning staff in particular, but

also comprised non-governmental and voluntary organisations. The DRDP opted for a low

input of expatriate personnel, namely one co-ordinator per district, formally operating in

an advisory position to the District Executive Director. Later, additional staff were added

in a few districts for specific tasks. The low input of expatriate staff was meant to enhance

the role and responsibilities of Tanzanian district officers, to create favourable conditions

for intensive popular participation, and to increase the sustainability of programme

results.

Most expatriate staff were employed by the Netherlands Ministry of Foreign Affairs and

aid was managed directly by the Netherlands Embassy (RNE). In three districts the

programme has been contracted out to the Netherlands Development Organisation SNV,

whose approach differs somewhat from the bilateral programme: it starts activities in one

division or sub-district within the selected district and gradually expands them to other

divisions. It provides more support staff, including a financial controller to improve

financial management, and a technical adviser for agricultural and environmental

conservation activities.

Initially, districts were selected on the basis of criteria such as type of farming system,

agricultural development potential, including possibilities for environmental conserva-

tion, development potential of rural smallholders, and access to markets. Later, other

criteria were also applied, such as location in the same region and seriousness of the

refugee problem.

The initial criteria reflected the emphasis on sustainable agricultural production growth,

which was also an important consideration in sector-oriented projects such as Farming

Systems Research and Smallholder Dairy Development. In the early 1990s, emphasis

shifted towards the social services, partly as a result of a change in Dutch development

co-operation policy, partly as a reaction to preferences of local politicians and govern-

ment departments in the districts, eager to share in the benefits of donor funding.

In addition, this shift corresponded with the tendency towards greater ownership in

Dutch aid. Also there was a link with Dutch project aid such as drinking water supply in

Shinyanga. Support to social sectors in DRDPs was further encouraged by the appoint-

ment of sector specialists on health and education in the Netherlands Embassy.

53



Project advisors in the various districts had a fairly high degree of autonomy in advising

about planning and expenditure, which generally led to a widening of the range of

activities supported and greater differentiation in the nature of programmes for indivi-

dual districts. In the mid-1990s, RNE started an attempt to streamline the programme in

the various districts. Regional co-ordinators were appointed, the district administration

rather than the expatriate co-ordinator became responsible for financial management,

Dutch funding became integrated in overall district development planning, and the

programme was made instrumental to the reform of local government. When the Local

Government Reform Programme began in earnest in 1999, the Netherlands signed an

agreement with Tanzania to facilitate the reform process. By the end of the 1990s the

Dutch aid programme for Tanzania was limited to four sectors, and DRDPs were incorpo-

rated into sector support to Local Governance.

The selection of local governance as a priority sector in Dutch aid to Tanzania had several

implications for the DRDPs. Firstly, it incorporated the support to individual districts in a

wider institutional framework encompassing macro- meso- and micro-level organisa-

tions. This is expected to facilitate linkages between local authorities and civil society, as

well as the effective use of resources geared towards local development. Second, the focus

has shifted from the implementation of activities for rural development to institution

building for a broad range of stakeholders. The central level is supported through partici-

pation in basket funding, technical assistance to the Ministry of Regional Administration

and Local Government, the reform team, and the Association of Local Authorities in

Tanzania (ALAT). At the meso-level, districts receive block grants to assist in the imple-

mentation of district development plans. At the micro or community level support is

directed towards the strengthening of civil society and people’s participation. Third, the

programme supports the funding of activities through District Development Plans, which

may enhance Tanzanian ownership of the programme. And fourth, the sector support to

local governance provides opportunities to integrate thematic priorities of Dutch develop-

ment co-operation policy (good governance, institutional development, environment,

gender and poverty alleviation) at the operational level.

DRDPs have frequently been evaluated: sometimes covering a single district, sometimes a

set of districts, and sometimes all districts supported under Dutch development aid.

In 1990, an evaluation mission concluded that little could yet be said about the effects of

DRDPs with regard to the increase of rural incomes and the improvement of rural living

conditions. The mission recommended that more attention be given to the main income



54

sources of the various districts and to priority for investments in the productive sectors

(CDP, 1991).

A subsequent evaluation in 1994 observed that DRDPs approached the poverty reduction

objective by giving priority to the creation of an enabling environment: improvement of

physical infrastructure and training of manpower. The report concluded that, although

DRDPs had a clear orientation towards the rural poor, they contributed only marginally to

the direct alleviation of poverty and to raising standards of living (IOV, 1994).

In yet another evaluation in 1999, the conclusion with regard of poverty alleviation was

that:

DRDPs reach only a limited number of poor villages and/or poor people, and in general there has

been no effort to identify the poor in particular and to target activities towards them. Road

rehabilitation has focused on potentially non-poor zones with relatively sizeable surpluses of

marketable crops, and other interventions have discriminated little among participants, apart from

those community development efforts directly channelled toward women(groups). Since the vast

majority of rural households is still considered poor, and taking into account the potentially more

relevant (but indirect) anti-poverty effects from rural growth, this non-discrimination is probably

less harmful than it may seem.

In addition, it was concluded that no real technological breakthrough had occurred, i.e.

sizeable yield increases in agriculture in line with development potential, and that the

rural population had benefited more from macro-economic policy changes than from

DRDP activities, although several activities were instrumental in capturing some of the

benefits, e.g. road rehabilitation and improved district planning (Bol, 1999).

The above evaluations concerned the four districts that had been included in the Dutch-

supported aid programme from the very beginning. In 2000, the effects of reforms

(including the Local Government Reform Programme of 1997), were evaluated for all

fourteen district programmes supported by the Netherlands. This mid-term review was

preceded by a so-called Facilitated Self-Assessment for each district. The picture that

emerges from the evaluation report can be summarised as follows:

• Democratic structures are in place, both political and administrative, but they have

been introduced from above. In almost all districts CCM has dominated the process

and councillors have been elected through the CCM party system, often unopposed.

These councillors chiefly promote party interests and are less ardent when it comes to

55



representing the interests of their electorate. Elements of the population with little

economic clout (e.g. women and the poor) usually have little political influence on

the village and district level.

• Staff performance is generally low, due to the lack of adequately qualified staff

combined with a lack of motivation due to poor working conditions, including low

salaries and limited career perspectives. Financial management is particularly weak:

only four districts have Principal Internal Auditors. External audits have been under-

taken as planned, confirming that the quality of financial management has generally

been poor.

• Formally, a participatory planning process has been followed. It appears that the

population particularly expects government to assist in building-up the service sector

and voices less demands for support for the production sector. There is still a high

degree of donor intervention in project selection and approval.

• Although there was an increase in overall council contributions to the District

Development Fund, in all cases the amount is still low in relation to total DDF expen-

diture. Most councils have major revenue raising problems, the collection of the

development levy in particular (RNE/SNV, 2000)

The report does not contain information with regard to the poverty reduction effects of

Dutch support to district development. It was noted, however, that the shift in emphasis

to capacity building for local governance may conflict with the now secondary objective of

providing a conducive environment for increasing rural incomes and improving the status

of women and other disadvantaged groups. In a joint reaction by RNE and SNV it is

emphasised that DRDPs have evolved largely into a capacity-building effort in the frame-

work of the Local Government Reform Programme The reaction re-iterates that “good

local governance is a major factor in creating an enabling environment for poverty

alleviation and as such will accelerate (rural) development in the districts.” It also

mentions the intention to raise expenditure on capacity building to the detriment of

direct poverty alleviation. (RNE/SNV, 2001).

In September 2003 the President’s Office/Ministry of Regional Administration and Local

Government and the Royal Netherlands Embassy commissioned a joint evaluation-cum-

formulation study for further support to the decentralisation process and development at

local level. The Terms of Reference of the mission indicate the nature of the support

envisaged. It should not take the shape of a project or programme but ‘’be fully

embedded in and compatible with the modus operandi of the Tanzanian Government as



56

evolving under LGRP”. This means that project-based provision of financial and technical

assistance is replaced by a funding modality at national level through regular govern-

mental channels, and in accordance with national policies as defined under PRSP and

LGRP. The Terms of Reference of the mission explicitly refer to the expected poverty

reduction effect of improved service delivery by local government institutions.


The massive foreign aid that Tanzania has received since the late 1960s has not resulted in

a substantial reduction of poverty, although almost all donors claimed poverty alleviation

as their main policy objective, and the high incidence of poverty (and sympathy for the

country’s policy) formed the main reason for including Tanzania as a priority country in

their aid programmes. There are two principal reasons for the low effectiveness of foreign

aid in terms of poverty alleviation. Firstly, most donors had no explicit poverty orientation

in the implementation of their aid programmes and substantial proportions of aid funds

were directed towards the transport and communication, energy and industrial sectors,

and primarily benefited the urban areas. Second, Tanzania’s development policy, and

especially the implementation of policy, was either bound to fail due to the strict state

control over economy and society exercised up to the mid-1980s, or was insufficiently

oriented towards the poor thereafter, at least up to the emergence of PRSP.

The low effectiveness applied to both dominant aid forms: project aid and macro-eco-

nomic support. In project aid, objectives were seldom expressed in terms of poverty

reduction and projects were generally too large and too complex and beyond Tanzania’s

managerial capacity. In addition, projects suffered from poor and hasty preparation, a

lack of adequate manpower, and often from insufficient government support. Moreover,

projects were executed in an unfavourable and sometimes discouraging macro-economic

environment. Several donors considered the multi-sector integrated rural development

programmes to be the main instrument for income generation and improvement of social

services in rural areas, but their effects towards poverty reduction were rather disappoint-

ing.

Macro-economic support, although less donor-driven than project aid, was geared

towards the construction of infrastructure and, in the industrial sector, towards a

relatively small number of private firms and parastatals located in the main towns, espe-

cially the capital. This aid had little effect on rural areas where most of the poor are living.

57



These characteristics also apply to Dutch development aid to Tanzania, most of which was

not directed towards direct poverty alleviation. Those activities that did focus on the

reduction of poverty were characterised by low effectiveness. The change in the Dutch aid

programme in the late-1990s towards a sector-wide approach, the selection of sectors

with a good potential for poverty reduction (education, health and local governance), and

the linking of sector aid to the objectives of the Poverty Reduction Strategy Paper, may

enhance its effectiveness for poverty reduction. These sectors do not fully correspond with

the priorities perceived by the population, however, and as yet there is no evidence of any

increased effectiveness.

Initially, DRDPs supported by the Netherlands had as their main objective to increase

farm incomes and to improve living conditions in rural areas. When they were incorpora-

ted into sector support to local governance, the focus shifted from the implementation of

activities for rural development towards institution building of a wide range of stakehol-

ders in local governance. Several evaluations undertaken during the 1990s mentioned the

low effectiveness of DRDPs in poverty reduction, mainly because little effort was made to

identify the poor, and to orient activities towards their main problems and the basic

causes of their poverty. In addition, activities were often small in scale; consequently,

there was low coverage of households and poor villages. The recent shift towards

institution building for local governance and the attempt to enhance Tanzanian owner-

ship may well have reduced further the attention given to income generation as a means

for poverty reduction.



58

7 Economic change and localgovernance in Mbulu district

7.1 District profile

Mbulu District is located in the northern part of Tanzania, and covers an area of 4,352

square kilometres. Until 1998, Mbulu district used to accommodate the present Karatu

district, and almost all available statistics include Karatu. In the census of 2002 the

Mbulu population totalled 238,000, i.e. excluding Karatu with a population of some

160,000. Population growth was 3.2% and average population density in Mbulu increased

from 60 persons per square kilometre in 1988 to around 90 in 2001. With 6.2 persons per

household the average household size is higher than elsewhere in the region.

The Iraqw dominate the district with around 90% of the total population, practising crop

cultivation and cattle husbandry. The Barbaig, pastoralists with minor (maize) cultiva-

tion, form only 8% of the district population, some 20,000 people. A few thousand

hunters and gatherers, the Hadzabe, complete the ethnic picture. From Irqwar Da’aw,

east of the present town of Mbulu, the Iraqw expanded in the past in all directions, stimu-

lated by an inheritance system that handed over all parental land to the youngest son.

Deviant social behaviour was punished by forced migration, with the aid of which tribal

elders maintained the social and moral order. Out-migration led to the ‘absorption’ of

other tribes, but without much violence.

Similar developments occurred in Karatu, the northern part of Mbulu District, but new

migrants included German and South African colonial farmers, who started large-scale

mechanised farming during the 1930s. These farmers later met with some resistance from

pastoral Barbaig and Maasai who feared the loss of their lands and rights, and their farms

were nationalised during the socialist period after 1967. The separation of Karatu from

Mbulu in 1998 resulted from a political struggle within the Iraqw, with Karatu more

wealthy than what remained of Mbulu district. Barbaig and Hadzabe have become

marginalised in the area, being without political influence, and their traditional way of

life has long been under serious threat.

59


Economic change and local governance in Mbulu district

Geographically, the district is characterised by extensive flat and gently undulating land-

scapes with hills and escarpments, and a high plateau in the East rising to 2200 metres.

There are substantial differences in land quality. Whereas Mbulu plateau has high rainfall

and good soil, the Lake Eyasi Basin is arid and rocky. Population density is consequently

very high on most of the plateau, surpassing 150 (per sq. km.); in the drier Western zone

near the Lake, unsuitable for crop cultivation, it is much lower. Temperatures vary with

altitude, averaging 20 to 25 degrees, but with very cold nights between May and July right

after the long rains. Rainfall ranges from 400 to over 1200 mm, with sharp annual

variations, but almost half the district receives less than 600 mm, the limit for reliable

agriculture.

Less than half the land area is arable; rainfed agriculture now occupies almost all arable

land and in many places there are signs of soil erosion. Protected forests cover about

72,000 ha or 10% of the district area. Apart from these reserves few forest resources

remain and degradation seems to continue, despite an official ban on all exploitation

since 1993. Illegal timber exploitation continues and a limited number of rural people

gain some income from the sale of firewood and charcoal. As a result of afforestation

campaigns in the 1980s, income from fruit (trees) and timber has increased.

Agriculture (and livestock) is the predominant source of income for 90% of the people.

There are no formal industries and the service sector is small. Some informal non-agricul-

tural activities include sorghum brewing, ghee making, mat making and tile and pottery

making. The introduction of electricity in some major towns in the year 2000 encouraged

non-agricultural activities, like welding, electrical contractors and shops for electrical

appliances and related goods. Trade liberalisation since the mid-1980s has also stimula-

ted various businesses.

Small subsistence farms characterise the agricultural sector. Average farm size was 1.4 ha

in 1995, but closer to 1 ha if the large wheat and coffee estates in the North, which is now

Karatu District, are excluded. Such estates vary in size between 80 and 1670 ha. with an

average of 425 ha. In most parts of the present Mbulu District there is increasing land

pressure; farmers and pastoralists compete for land, and landlessness and the out-migra-

tion of young people occur in several areas. Land rental – previously unknown – now

occurs in parts of the district. Food crops include maize, beans, millet, wheat and

potatoes.



60

Agriculture in Mbulu is a chancy business due to the heavy dependence on rains, wildlife

damage and the frequency of disease. Consequently, land productivity is low and even

declining, due to over-exploitation and lack of soil care (no fertiliser use). The people of

Mbulu face periodic food shortages and child malnutrition is still prevalent. Only in the

new Karatu district, with its good rains and soils, is the situation different and more

promising as a result of cash crops (wheat and coffee), some irrigated horticulture and a

good location near main roads and markets (Arusha). At present, Karatu is the second

61



Map 2 Mbulu District

largest wheat producing area in the country, and small farmers increasingly supplement

the large output of commercial (and mechanised) farms in the area.

Livestock is also important in Mbulu/Karatu, with numbers of main types (cows, goats,

chickens) remarkably similar to the population (400,000). Yet about a third of the popula-

tion do not own cattle, usually a sign of absolute poverty. In relation to population pres-

sure traditional livestock systems are under threat; the shortage of pastureland, increased

livestock disease, frequent drought and inadequate veterinary services are all debilitating

factors. Traditional livestock systems have only marginally been replaced by modern dairy

farming.

7.2 Economic change

The economic dynamics of Mbulu/Karatu are linked primarily to the territorial expansion

of the Iraqw and their role in wheat cultivation in Karatu. Up to the end of the German

colonial period, the expansion did not cause any basic changes to the rural economy,

which was characterised by an agro-pastoral production system and self-sufficient house-

holds. Before the 1920s, the staple crop in most villages was millet combined with

sorghum, maize, beans and sweet potatoes. Thereafter, maize replaced millet as the

staple food crop and Irish potatoes were also introduced. Cash income was obtained by

the sale of livestock and incidental surpluses in crop production. After the 1940s, the

British administration encouraged the production of coffee and pyrethrum as cash crops.

In the early 1940s, Iraqw migrants followed the example of German and South African

farmers and with their support and that of the British administration, started commercial

wheat growing. After a temporary slow-down in the 1950s, migrant/farmers again showed

a strong expansion in the first half of the 1960s. By the middle of that decade when all

suitable land in Mbulu/Karatu had been occupied, the Tanzanian Government opened-up

a new wheat-growing area in neighbouring Hanang district where most expansion

occurred thereafter (supported by Canadian aid).

The production of coffee and pyrethrum in the eastern highlands was expanded from the

late 1950s onwards. Maize surpluses were sold, initially through a Grain Storage

Department established in 1949, but later through co-operatives. The 1960s was thus a

period of increased cash crop production and rising incomes.

Villagisation policies in the 1970s had little impact in the densely-populated higher

areas, where most habitat and land use remained as it was, but they frustrated life for the

pastoralists and their animal husbandry systems in the Western zone.



62

Although the first years of ujamaa did not yield any improvement in agricultural produc-

tion, rural living conditions did not deteriorate. The government abolished local taxes,

and afterwards primary school fees (in 1973), and initiated generous programmes by

which to expand the social infrastructure in rural areas, notably water supplies, schools

and health centres. The peasants in Mbulu were also spared the heaviest burdens of

socialist transformation; hiding behind the peculiarities of their area (land pressure),

they were able to feed themselves and to earn some cash, although socialist rhetoric and

talk of communal farming by the authorities complicated their work. The call to intensifi-

cation with the manifesto Siasa ni kilimo (politics is agriculture) in 1972 met with limited

response in Mbulu, nor was there any indication that peasants produced more in

response to government generosity with social welfare measures (Chachage, 2001).

The situation changed during the 1970s, when low producer prices and the mishandling

of produce by the local co-operative society caused coffee and pyrethrum production to

decline. Prices for food crops were more favourable, however, and production grew with

the expansion of the farming population. The area under cultivation increased, but at the

expense of land of the pastoralists.

The marketing of agricultural produce was liberalised during the late 1980s. Initially, the

Arusha Regional Co-operative Union and the Rift Valley Co-operative Union and their

primary societies maintained a quasi-monopoly position, but their functioning was

problematic due to liquidity problems, poor management and some corruption.

Given their financial problems, the co-operatives were no longer considered credible

customers by the banks and lending arrangements were no longer honoured.

The co-operatives also faced strong competition for food crops from private buyers, who

bought the crop at the farm gate and paid instant cash. Moreover, prices paid by private

traders were often lower than the official indicative prices.

Although liberalisation created some benefits in freer trade and new private business, it

did not help agriculture very much as the market by itself was unable to compensate for

the loss of important public services in extension, input supply and marketing. Maize

production, the main food crop, fluctuated strongly during the 1980s, i.e. between 60,000

and 100,000 tonnes with a slightly upward trend, but has fallen since the end of the

decade. Other food crops also seem to have decreased during the 1990s, although

available figures are only educated guesses and adverse weather conditions may have

biased the annual comparisons.

63



In general, yields are stagnating or declining as a result of soil erosion and environmental

degradation, and annual variations are high (NEI, 1997). Since 1989 in particular,

declining maize yields have been compensated by increasing the area under cultivation.

Yields depend very much on soil type and rainfall conditions, and thereby differ regionally

as well. Maize yields, for example, varied from 0.7 to 3.7 ton/ha in the mid-1990s, with the

highest yields being recorded in the North (now Karatu) and the lowest in the poorer East,

although substantial differences in yield levels occurred among farmers even within these

two areas (NEI, 1997).

Low yields can only be overcome on irrigated plots, doubling the above values on average,

certainly when high yielding varieties and farm inputs are applied. The capital required

for this type of agriculture usually restricts it to the better-off farmers. Only 1% of the area

is at present under irrigation, mostly in Karatu, and there is little scope for expansion.

For some villages irrigated plots provide a sizeable income through a marketable surplus,

including horticultural crops such as onions and garlic. Marketing remains a bottleneck,

however.

7.3 Social change

Social developments in Mbulu have more or less followed national trends, i.e. a rapid

increase in social services during the 1970s, followed by a decline in the 1980s, and

without many signs of recent improvement. The enrolment rate for primary education

dropped from around 90% in the 1970s to a mere 67% in 1995, with equal numbers of

girls and boys, only few of whom – 13% in 2000 – continue with secondary education.

The infrastructure is inadequate and has been so for many years. There is an increasing

lack of classrooms (15% short of required capacity), teachers’ houses (48% short), as well



64

Table 7.1 Production of main food crops 1991–1997, including Karatu (metric tonnes)

Crop 1991/1992 1996/97

Beans 12,000 7,000

Maize 65,000 55,000

Millet 12,000 5,000

Wheat 15,000 14,000

Source: District Office, Department of Agriculture and Livestock Development, 2000

as desks and equipment. On average, 75 pupils occupy one classroom, and 13 have to

share a book each (MDC, 2001).

Health care infrastructure is in accordance with national standards. There are two hospi-

tals in the district, a public and a private one, 24 dispensaries, 4 primary health care units

and 36 mobile clinics. This means that in general the population to be served by one

facility comes close to national standards of 10,000 per dispensary and 50,000 per health

centre. Facilities are unequally distributed over the district, however, and there is little

information about the quality of the services and access to them on the part of the poor.

A little over 50% of the population has access to water schemes, chiefly shallow wells.

Almost all shallow wells and boreholes are in working condition, but of the eight piped

water schemes, only four are in operation. Safe and clean water is accessible to 43% of the

population (MDC, 2001).

Social relations are also changing as a result of economic changes over recent decades

and the apparent stagnation since the early 1980s. Land shortage has affected traditional

family relations, undermining the authority of male elders. Increasingly, male heads of

households are losing control over the affairs of their sons/daughters and wives. Male

involvement in farming has declined, contributing to pronounced alcoholism amongst

adult males. Younger men, women and girls increasingly work on their own individual

economic ventures, including growing crops (fruits) on rented land, petty trade, pig

raising, charcoal making/selling, brick making and stone crushing, and running shops.

Some of the younger people leave home to seek work in the towns (Chachage, 2001).

Traditional society and its values have increasingly been eroded by contemporary forces,

such as conversion to Christianity, modern education and ‘development’. Associated with

the outside world, such phenomena are viewed by older people as a potential threat to the

community, to Iraqw culture and traditions. Younger people, on the other hand, regard

them as a means to become a modern Tanzanian, gaining access to new knowledge and

skills, and acquiring a superior status. To the young progress means breaking with the

past, and thus also rejecting farming as an economic activity and a ‘primitive’ way of life.

The economic possibilities of farming, at least those of subsistence farming, do not look

very promising to the young, as it fails to provide the monetary surplus that could be used

to acquire of what they see as symbols of development: western clothes, cooking utensils

and corrugated iron roof sheets, radios, bicycles, cars and tractors. But there seems to be

65



no alternative: as a result there is a social crisis among the young, with much theft and

drunkenness; rejection of farming and moving to towns, where they find few viable

alternative income opportunities (Snyder, 1993).

7.4 Local governance

Before the arrival of German colonial officials in 1905/06, the various ethnic groups in

what is now Mbulu/Karatu were organised as separate socio-cultural and political enti-

ties, without any clear state structure. The Germans tried to co-opt some local leaders

into their administration as headmen, but this failed due to conflicts over taxation.

When the British Administration took over in 1916, they established a formal government

structure and, through the Native Authorities Ordinance of 1926, a District Authority for

Mbulu was created with an Iraqw chief. Gradually, three other ethnic groups had their

chiefs formally appointed, presiding over an ethnic Native Council. The chief and sub-

chiefs functioned both as political leaders and justice administrators, maintaining law

and order, ensuring efficient tax collection, and mobilising people for public works.

In the 1950s, a district council was established with appointed members only. Although it

was not a representative body, its creation provided an avenue for the growth of a local

political elite, active in the nationalist movement of the late 1950s. On the attainment of

Independence, Mbulu local authority was abolished and the District Council was re-

organised to enhance the role of nationalists. Membership included elected individuals

and appointees of central government. With the abolition of the multi-party system in

1965, the only political party (TANU; later CCM) became part of the administrative

structure, acting as a shadow government in Mbulu as elsewhere in the country.

The changes made to the system of local governance in 1972 were basically a further

centralisation of power and authority. A hierarchy of administrators and planners ran

from the regional level via district and ward down to the village level. At all levels, officials

combined party and government authority. Local governance became a system by which

to channel directives from the top downwards, an avenue for receiving directives from

higher level authorities, such as the nation-wide villagisation campaign in 1974/75.

In practice, the centralised system continued in Mbulu up to the early 1990s. Although a

process of renewed decentralisation began in 1982 and District Councils were re-intro-

duced, no major structural alterations were effected, and at both the village and district

level the government administrative structure remained tied to the single-party machi-

nery. The re-introduction of the multi-party system in 1992 necessitated the separation of



66

party and government functions at the village, ward and district level. As long as CCM

ruled, few difficulties arose, but when the opposition won in Karatu in 1999, the local

government structure became part of the political struggle. Some civil servants were

removed; both sides tried to control the area, taking advantage of the ill-defined respon-

sibilities, rights and duties of the new local authorities vis-à-vis central government, at a

time of structural reform. Although decentralisation and more local governance has

officially been on the agenda since 1999, in practice little has changed, partly due to

resistance to change by powerful elites and partly to lack of capacity of local govern-

ment(s) to take advantage of the new opportunities. A new element is the growing

influence of NGOs and other grassroots civil society organisations, in the wealthier

Karatu more than in Mbulu. Thus far, such private initiatives have more or less been

tolerated by the authorities but with little enthusiasm, and authorities and NGOs work

alongside one another without much interaction or co-operation.

7.5 Perceptions and poverty

In the mid 1990s, i.e. when Mbulu included Karatu, average rural household incomes

were estimated at around US$ 1500, or almost US$ 250 per capita (NEI, 1997). District

variations were large, however, with farmers in (small) irrigated areas reaching almost

double that average, and those in the Eastern Zone – including pastoralists – hardly a

third. The averages were also based on reasonable rainfall, which has not occurred for a

number of years in recent decades. On average, livestock generates one-third of total

income in the district and non-agricultural activities only 5%. If poor peasants - mostly in

the eastern Zone – are added to those households without any livestock, then 40 to 50%

of the population is poor in an absolute sense, i.e. earning less than a minimum with

which to satisfy basic needs.

Perceptions on progress and poverty differ among the three ethnic groups in the present

Mbulu district. The small group of Hadzabe hunters and gatherers experience hardly any

social or material differentiation and lack any material goods or livestock. They distin-

guish between those who own nothing (i.e. the Hadzabe themselves), and those who own

complex things such as clothing, cattle, cars (i.e. all others). Property has no material

meaning to them and they criticise those who want to possess more than is strictly

necessary but do not want to share anything. The outside world, and development,

typically represents this greed, which they consider as threatening to their potential

wealth (game) and way of life (Stephenson, 2000).

67



The Barbaig link poverty with social independence: a poor person is someone who has to

be helped by others and in turn is unable to help others. A poor person does not have any

(dairy) cattle and therefore cannot feed or socially reproduce himself, thus making the

person insignificant, also in social terms. A wealthy person has many cattle and is

respected for his success and wealth. To be respected, however, also involves the social

obligation to give or to lend cattle to the poor, so that they are helped or can help

themselves. Wisdom and leadership are also expected from a rich man, to increase the

well-being of the group.

In Wa-iraq, poverty refers to the lack of food. Having enough to eat and (as a result)

ensuring harmony in society is associated with progress. From this point of view, periods

of poverty regularly occur in Mbulu district, as in 2001 during the present study. Iraqw

society has different mechanisms for lending cattle to families who suffer from lack of

food, but those mechanisms are under pressure due to the declining number of cattle,

which is partly caused by the practices of tax collectors. Among the Iraqw, poverty also

has a social dimension, and deteriorating relations (within the family, between neigh-

bours, in the community), associated with stagnating food production are perceived as

poverty. The term development, on the other hand, is not really associated with food

production, growing herds or material possessions, but rather with the mentality that is

coupled with the modern (post-colonial) time: people who belong to a Western religious

community and have a Western education, i.e. “people of the pen”. The young people

believe that this category contrasts with illiterate farmers, who are associated with

old-fashioned ideas and poverty. Development has now lost some of its appeal due to the

negative connotation with the corruption of leaders, which has hampered the progress of

ordinary people.

Irrespective of ethnicity, the people’s perceptions of poverty differ from those of the

government staff and leadership of the district. The latter perceive poverty in economic

terms, in the lack of agricultural output, food shortages and low food security, and low

cash incomes.

Many villagers attribute increasing poverty to the fact that government seems to have

relinquished its supervision over agricultural and livestock sectors. While government

shaped the agricultural sector compulsorily in the past, it has neglected it in the recent

years, particularly agricultural extension and veterinary services. Both farmers and herds-

men state that crop and cattle diseases have increased of late, the market situation is

deteriorating, and the private sector fails to take over the former role of government.



68

They complain about the lack of inputs (suitable seeds, fertiliser, pesticides). and they

consider the ongoing privatisation process as a conspiracy between government and

private investors.

For example the cattle diseases. Our leaders said they are going to work on it. They know but they

haven’t done anything until now. Our cattle vanished. Nothing is left. The medicine that we buy is

our own effort to cure the diseases, but it kills the cattle. They help us nothing actually. If any

assistance would have been provided to cure at least half the cattle that would have been a

starting point (video).

Also perceptions about the quality of social services are quite unfavourable.

Our dispensaries are a problem. They say treatment is for free if you pay 10,000 shilling.

But when you pay they say there is no medicine for treatment. Some people don’t pay and die at

home, because if you pay, you only get aspirin and you will die anyhow. Our leaders tell us, this is

the normal thing.

And on education:

Today everything is costing. Parents have to pay levy and school fees and if you cannot pay

school fees you sell a goat. But in this land we depend on crops and if the crops fail we sell a goat

for food…. We have been building schools for nothing. Children are staying outside, teachers are

not teaching, they only look after their salaries. People’s education has gone down. What is

development? (video).

Perceptions differ amongst local people about the quality of local governance. All note a

lack of transparency in development planning and the flow of financial resources caused

by the hierarchical top-down process, but many people (especially the older) state that:

Planning is done by our village leaders, then the plan is explained to the villagers in a public

meeting. Once the leaders have prepared a plan, it cannot be rejected by the villagers.

The young increasingly seem to challenge such obedience. They complain about the

corruption of leaders and associate corruption and paternalistic tendencies among

present-day leaders with their literacy and ability to communicate in the bureaucratic

language. They also complain about the functioning of courts of law, which they consider

unpredictable and outside their influence, and about the lack of communication between

leaders and ordinary citizens (Chachage, 2001).

69




Mbulu District has an unfavourable resource position: in large areas only half the land is

arable, soil fertility is low and rainfall erratic. Only the Mbulu plateau has good soils,

high rainfall and a high population density. This formerly largely coincided with the

Karatu Division and is the present Karatu District.

The district economy is dominated by agriculture; the main source of income for 90% of

the population. There are no secondary industries and only a small service sector. Most

farms are small in size and land productivity is low. Moreover, due to over-exploitation

and low input use, land productivity shows a declining trend. Many farmers produce

barely enough to feed their own households, and commercial crops are fairly insignifi-

cant. The exception is the new Karatu District, where farm incomes are substantially

higher due to commercial wheat and coffee production and irrigated horticulture.

Mbulu district has a large livestock population but the traditional animal husbandry

system is characterised by low productivity and one-third of the households do not own

any cattle. Moreover, farmers stated that recently they were forced to sell livestock in

order to pay their development levy.

Rural economic change has been associated closely with the expansion of mechanised

wheat farming and coffee growing. Since the 1940s and especially during the period

1950-1965 wheat production expanded spectacularly as local Iraqw migrants adopted the

mechanised farming system. Outside Karatu, there was some commercialisation of

agriculture (coffee and pyrethrum), but in the 1970s low producer prices and mishandling

of produce by the co-operatives caused a decline in output and farm incomes. Economic

reform did not bring any substantial improvements to production conditions as the

market did not compensate for the loss of important public services.

Social change involved an increase in social differentiation in relation to wheat cultiva-

tion and a shift of powers to certain groups of Iraqw, and the marginalisation of other

ethnic groups. As elsewhere in Tanzania, there was a rapid increase in social services

during the 1970s, but these declined in magnitude and quality in the 1980s. The present

situation in education is characterised by a serious shortage of classrooms, teachers’

houses and teaching materials. Health care infrastructure compared to population is in

accordance with national standards, but there is little information about the quality of

health services.



70

Social relations are changing as a result of increasing economic differentiation and the

stagnation of the rural economy since the 1980s: this has undermined the authority of the

male elders and contributed to pronounced alcoholism among adult males. Most young

people associate progress with breaking with tradition, and reject farming as a ‘primitive’

way of life. The lack of alternative economic opportunities, however, causes social crisis

among them.

After Independence people in Mbulu experienced a short period of improved living

conditions; with its present boundaries, however, Mbulu is a district with a high degree of

poverty due to low incomes and deficient social services. An estimated 40-50% of the

population is poor in an absolute sense, i.e. earning less than a minimum to satisfy basic

needs. There are periodic famines and the majority of people express poverty in terms of

food shortages and deteriorating social relations. They relate increasing poverty with

government’s neglect of agro-support services and the weak performance of the district

administration. People’s perceptions differ from those of government staff, who see

poverty in narrow economic terms: low agricultural output, food security and incomes.

Perception studies indicate little confidence that government will help to solve the

problems. The young in particular complain about the corruption of leaders and the lack

of communication between leaders and ordinary citizens. There seems to be deadlock

between government and citizens, between leaders and ordinary people about the strate-

gy for development and poverty reduction in an area with a low potential for agricultural

growth, high dependence on agriculture and deteriorating social relations.

71




Lessons for the HIPC process

72

8 District rural developmentprogramme in Mbulu

8.1 Introduction

The Mbulu District Rural Development Project (MDRDP) was the first district programme

supported by the Netherlands in Tanzania in 1987. After a pilot phase lasting two years, in

which some starter activities were identified and initiated, a second phase took place in

1992-1996 and the programme is now in its third phase (1997-2003). When Karatu

became a separate district in 1998, a DRDP was also set-up for the new district, becoming

operational in 1999/2000.

During the pilot phase, a process approach was followed and a strategy developed on a

trial-and-error basis using so-called starter activities (which included the rehabilitation of

roads, agro-forestry, repair of buildings and training of government staff ). Basic data

were collected through a baseline survey, but was considered to be of doubtful quality

and little use was made of it. While the emphasis in the district programme was on

income-generating activities, no detailed poverty analysis was made, nor any distinction

among various population groups apart from the gender differentiation. General deve-

lopment planning by various government levels, from District Council to Village Council,

economic infrastructure (mainly roads), environmental issues and rural water supply

were identified as key areas for subsequent intervention.

8.2 Objectives and strategy

The long-term objective of MDRDP has always been to improve the standards of living of

the rural population in a sustainable or structural way, be it that incomes were central in

the first years, For the second phase (1992-96) objectives were reformulated, allowing for

a somewhat broader approach, emphasising standards of living rather than incomes

alone. This made it possible to include social sector support from then onwards, explicitly

excluded in the first phase. One of the reasons for this kind of support was to try to safe-

guard Tanzania’s social achievements by increasing incomes and thereby facilitating tax

collection for maintaining education and health services. Environmental concerns were

also addressed from then onwards. During the third phase (1997-2003), good governance

was explicitly added as a major concern, when the Netherlands re-oriented its assistance

through sector support, whereby DRDPs became part of the sector ‘local governance’.

73


District rural development programme in Mbulu

Since the start of the third phase in 1997, the overall objective of MDRDP has been to

improve the living standards of the rural population in a sustainable way, by topping-up

the development budget of the District Council and by institutional strengthening of the

public and private sectors (MDC, 2001).

This overall objective has been further specified in four so-called project purposes, i.e. to

improve people’s income; health status; educational standards; and to strengthen the

management of development initiatives and projects through capacity building at all

administrative levels (good governance).

Furthermore, each ‘purpose’ has its specific objective and priorities, from which clusters

of activities have been derived (see 8.6). All objectives, purposes and activities have been

brought together in a Logical Framework, which in principle allows for a proper monitor-

ing of outputs and results, as well as a check on underlying assumptions. Additional

criteria for the selection of activities were developed to do justice to the objectives of

Dutch development co-operation, such as poverty alleviation, participatory planning,

gender and environmental issues.

From the start, the main characteristic of the strategy has been strengthening of the

District Council. Co-operating closely with the District Executive Director (DED), the head

of all district council activities, and the Planning Office, technical and financial

assistance was provided by the Netherlands to improve the planning process of the

Council and its departments, at first mainly of its productive activities and then also of

social sectors.

Planning at all levels including village and ward had early been identified as weak and

top-down, with too much influence by the Council’s technical staff, whose qualifications

and work experience left much to be desired. Little attention had been given to people’s

priorities expressed through their official representatives in village, ward and district

committees and councils. Major efforts were put into the training and capacity building

of heads of departments and other civil servants, in order to improve the participatory

nature of official development planning in rural Mbulu. A great deal of attention has also

been given to improving the basic planning process at the village level, in particular

financial monitoring. Village development facilitation teams have been deployed to

reinforce the participatory development process in the villages. In 1999 ward level

planning was introduced, replacing the time-consuming village planning exercise, while

it was attempted to enhance popular participation. Recently, NGOs and the private sector

have been approached to assist in implementation of the plans.



74

Some starter activities selected in the initial years continued throughout MDRDP’s history

to encourage the required changes in the planning process. Much training and technical

assistance was organised around these activities, with a constant struggle with the politi-

cal leaders to balance hardware (construction) and software (training). However, develop-

ment planning and strengthening of the institutional planning framework became the

central orientation of the programme from the early 1990s onwards, with efforts aimed at

a qualitative improvement of plans, budgets and implementation, strictly supervised by

the programme co-ordinator. Activities funded by the Netherlands were brought together

in clusters, first land and water management, later followed by health and education.

At the end of the third phase, MDRDP is basically a financial contribution towards the

District Development Fund, and towards audits at the Council level, with diminishing or

completely disappearing technical assistance. How this will influence the planning

process at district level, and the utilisation of Dutch funds, remains unclear.

8.3 Programme organisation and finances

Institutionally, MDRDP functions through the District Council. The executive arm of the

council, i.e. its civil service, designs all development plans and is headed by the District

Executive Director (DED). The DED is in charge of the district team of 12 Heads of

Department and experts engaged in the planning and implementation of activities.

Although these departments formally fall under Mbulu District Council, a number also

receive instructions and funds from Central Government. This means that they partially

escape the supervision of the District Council and its elected Councillors. In 2001, Mbulu

District Council had a staff of 1070, over 600 being primary school teachers.

The Dutch co-ordinator played an important role in shaping MDRDP, its strategy and

budgetary allocations, in co-operation with the District Planning Office under the

guidance of the DED. The DED and the Dutch co-ordinator were both to sign for all expen-

diture. In the early stages there was more technical assistance provided by a permanent

(expatriate) water engineer and a part-time road engineer. Both functions were abolished

in the second phase, and from then on only short-term consultants were hired from

outside, usually Tanzanians, except for fairly intensive and recurring assistance from the

Netherlands in shaping the land and water management programme component.

The Dutch programme co-ordinator also maintained close contact with heads of depart-

ment at district level and their staff, and with individual ward or village managers and

75



farmers. Relationships with councillors have always been somewhat marginal, reflecting

some tension between politicians and executive branches.

During the third phase more integration of MDRDP into the District Council is aimed at,

with the Dutch co-ordinator becoming more of an advisor, especially since Dutch funds

are now directly channelled to the Council, be it surrounded by strict audit control.

During the period 1988-2002, the Dutch contribution to MDRDP totalled almost US$ 8.5

million. Aid allocations gradually increased from US$ 392,000 in 1988 to US$ 855,000 in

1996. The subsequent decline was caused chiefly by the separation of Karatu into a new

district, with its own DRDP. The new Karatu DRDP received US$ 374,000 in 1999 and US$

369,000 in 2000. Disbursement levels were also influenced by the absorption capacity of

the Council, and the ability of its staff and people to take care of it properly. Budgeted aid

was never fully spent, and less so in the late 1990s when attention shifted more towards

the maintenance of earlier investment in education, health, roads and water. Moreover,

RNE released funds in accordance with the matching principle linking Dutch aid to coun-

cil investment funds. This often meant that council had to increase revenue and intensify

local tax collection.

Table 8.1 shows expenditure by the Netherlands on MDRDP throughout its lifetime.



76

Table 8.1 MDRDP expenditure 1988–2002 (in '000 US$)*

Year Expenditure Year Expenditure

1988 392 1996 855

1989 315 1997 764

1990 546 1998 448

1991 572 1999 456

1992 410 2000 416

1993 421 2001 590

1994 822 2002 703

1995 784

* up to 1999 including Karatu

Source: Various Evaluation Reports (1990, 1994, 2000), RNE files

Not only disbursement levels but also the distribution of funds over the main sectors has

varied considerably over the years. Unfortunately, no comparable sector allocation is

available for the whole period. In the early years only domestic (local) expenditure by the

Netherlands could be traced per sector; this excluded all foreign expenditure, imported

road equipment, materials and foreign consultants, i.e. the majority of expenditure

certainly in the early stages of the programme. With regard to local expenditure, a shift

occurred between the period 1988-89 and 1991-93 as Table 8.2 shows.

It is clear that relative expenditure shifted mainly from economic infrastructure (roads,

go-downs) to social sectors (health and water at first). In absolute terms, however, this

caused little repercussion as total domestic expenditure by the Netherlands increased,

from around US$ 0.9 million (Tsh 110m) in 1988/89 to US$ 2.1 million (Tsh 650m) in

1991/93.

The sector distribution of expenditure is only available from 1997 onwards, and is

reflected for main sectors in Table 8.3.

After heavy investment in roads and agriculture at the start of the programme, in the mid-

1990s attention shifted towards the social services sector, especially education. Recently,

this sector has received about 40% of all expenditure. Renewed efforts to rehabilitate the

economic infrastructure, especially roads, are again becoming visible, while the Council

is spending much less on directly productive activities as in agriculture. This is in line

with the new strategy explained above, even though expenditure alone can never tell the

whole story. Still, it is remarkable that during the period 1997-2002 only 10% of all Dutch

support went to agriculture (plus 3% to livestock), as compared to 60% to the social

77



Table 8.2 Sector Distribution Domestic Expenditure MDRDP, 1988-1993 (%)

Sector 1988-89 1991-93

Overheads 30 20

Agriculture 15 20

Roads 30 15

Go-Downs 15 5

Social services 10 40

Source: Evaluation Reports 1990 and 1994, CDP

sectors (mainly education). In 1999 a large investment was made in the participatory

planning process, i.e. 22% of the Dutch contribution in the ‘supporting sector’ (admini-

stration), but this percentage has decreased substantially again in later years.

Similarly, since the start of the new DRDP in Karatu in 1999, only 6% of expenditure has

been devoted to agriculture and livestock, and 54% to the social sectors (Annual Report

2001, RECO).

It is possible to reconstruct broadly the long-term trend in DRDP development expendi-

ture over main sectors on the basis of tables 8.2 and 8.3. This trend is presented in table

8.4 and clearly reflects the shift from productive activities to social services.

The financial contribution by the Netherlands is only part of the total (development and

recurrent) budget for the district. About half the total district budget comes from central

government (nearly all recurrent costs, mainly salaries, and a minor part of development

costs). Only 8% of the total district budget comes from tax revenues. The picture changes

if the analysis is limited to the development budget, which largely indicates investments.

In the late 1990s MDRDP represented about 80% of the development budget without



78

Table 8.3 MDRDP sector distribution for development expenditure, 1997–2002 (%)

Sector *1997 *1998 1999 2000 2001 2002 1997/2002

Education 35.9 22.7 34.6 49.0 40.7 46.6 38.8

Health 7.1 8.6 6.2 6.3 6.6 5.1 6.5

Water 13.2 22.3 5.2 5.2 26.6 13.4 14.6

Social services 56.2 53.6 46.0 60.5 73.9 65.1 59.9

Agriculture/ Land use 19.9 16.8 15.1 18.5 15.5 15.0 16.8

Roads/ transport 8.3 19.8 15.9 13.2 3.0 13.2 11.6

Supporting sector 15.3 9.3 22.4 7.1 6.8 5.4 10.9

Other 0.3 0.7 0.6 0.7 0.8 1.3 0.8

Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0

* including Karatu

Source: Annual Reports RECO

counting the own contribution of the population. In principle, the people were supposed

to contribute 25% of the development cost of their village projects. Such self-help (in

cash and kind) is often considerable, e.g. when investing in roads, schools and clinics;

including such help might reduce the overall Dutch contribution to the Mbulu develop-

ment process to around 60%.

In the third phase of DRDP a formal matching principle was introduced, linking Dutch

investment cost to total council revenue. As a result, tax collection by Mbulu Council

increased from 1998 onwards and, apart from its regular contributions to recurrent cost

(mainly staff salaries), its contribution to the development budget increased from US$

33,000 in 1998 to US$ 44,000 in 2000. This is remarkable in view of its zero contribution

in 1997 and US$ 8,000 in 1996 (MDC, 2001).

Villagers are also supposed to contribute more to projects as a result of the matching

principle, in some cases up to 50% of development cost. This target has proved hard to

achieve in addition to the maintenance of schools, health units, roads, bridges and water

systems which also has to be taken care of by the people.

8.4 Activities and output

Since 1987 MDRDP has supported and financed a great number of activities. In spite of

broad shifts towards social sectors and subsequently to strengthening local governance,

many activities started in the initial period of the programme still continue in some form

or other. The 2001 Plan lists 66 activities for the period 1998-2000, combined into nine

clusters, of which five under income generation (26 activities), two under health (15 acti-

vities), one under education (11 activities) and one under institutional capacity building

(14 activities). Table 3.2 in Annex 3 summarises major activities.

79



Table 8.4 Trends in financial assistance, Mbulu DRDP, 1988-2002 (%)

Sector 1988/1989 1991/1993 1997/2002

Agriculture 40 30 17

Roads / Transport 45 20 12

Social services 15 50 60

Supporting sector – – 11

TOTAL 100 100 100

Activities under income generation have dealt with the promotion of crop production and

livestock, environmental protection, group formation amongst farmers (savings, credit)

and road rehabilitation. Improved health services are aimed at through the training of

government staff at district and village levels, while the reduction and/or prevention of

disease is promoted through health education and awareness campaigns amongst the

village people, as well as some improved infrastructure, like a cold chain for drugs, water

wells and tanks. Education is being improved through teacher training, in common

subjects like languages and mathematics, as well as in specific topics such as gender,

environment, health education and accounting. Infrastructural improvement is also part

of the programme, including new or rehabilitated classrooms, schools and/or teacher

houses.

Finally, the capacity of the District Council is being improved through much training of

its staff at district, ward and village levels and in various subjects (accounting, computer,

planning, gender, land use and maintenance), and the sensitisation of communities in

participatory planning, extension and development.

The number of activities per sector indicates that the allocation of (aid) funds over sectors

does not always provide an accurate picture of activities supported by Dutch assistance.

The income-generation cluster still includes most activities. With the exception of road

rehabilitation, many are apparently small-scale and labour-intensive, less capital-inten-

sive than the rehabilitation of schools and clinics.

It is only in the third phase that MDRDP output has started to be recorded systematically.

The District Plan 2001 provides comprehensive information on output during the period

1998-2000 in relation to the activities discussed above. A summary of output for that

period is provided in table 3.3 in Annex 3. In addition, there is some indication of output

prior to 1994 (CDP, 1994). Despite the gap (1994-1998) in output data, the continuity in

types of activities and output throughout the 1990s is considerable, with some increase in

numbers of people reached towards the end of the decade, but not in all activities.

In agriculture the focus continued to be on training farmers in new production methods

(improved seeds, disease control), new technology (oxen, irrigation, storage), new crops

(sunflower, pyrethrum, fruits, vegetables), and encouraging some group formation in

order to bargain collectively for inputs and credit. The most prominent improvement was

that of the adoption of high-yielding maize varieties. This adoption increased to half of all



80

farmers in two villages, with a yield difference of at least 50% between project and non-

project farmers, and the introduction of coffee growing by 10% of farmers. In livestock

improved dairy cows were introduced to boost milk output, some training in pig hus-

bandry has been provided, and attention given to the improvement of traditional herds

(dipping, disease control, spraying). Many of these activities were not new in themselves,

and were undertaken by government staff, but DRDP aimed at improving the perfor-

mance and success of such public services to farmers and thus funded their implementa-

tion. The programme also financed the provision of a limited number of water and

storage structures and oxcarts, and thousands of new (fruit) seedlings. New savings and

credit societies replaced the earlier co-operatives. The various agricultural programmes

reached an estimated 7500 farm households throughout Mbulu District.

In natural resources protection, an intensive land and water management programme

was introduced in 1993, including village land-use planning, conservation of farmland

and protection of water sources, distribution of seedlings for agro-forestry, and farmers’

training in soil management.

Support to education started in 1991 with the construction and rehabilitation of class-

rooms (about 225), teacher houses (140) and pit latrines at schools (33). Since 1994 the

training of teachers (over 500) and other methods to improve the quality of education

(pupils pass marks, school inspection) have received more attention. A health pro-

gramme had also started in 1991, since when about 1000 staff had received training,

mainly traditional birth attendants in villages. In addition, dispensaries have been

renovated and newly equipped (radios, refrigerators, water tanks/wells), and small

groups of local people trained in preventive health care.

In the area of institutional changes and capacity building, DRDP reports an increase of

revenue collection by the District Council, more job training for government staff, more

community facilities, positive changes in attitude towards participatory development and

knowledge on gender among government staff, and improved accounting systems.

Despite the numbers of villages and people supported by DRDP activities, the proportion

of the population included in the programme is rather low. Five types of activities reached

a coverage of 25% of all villages in Mbulu in 2000 after some 10 years of intervention. These

include soil conservation and agro-forestry, agricultural extension, dairy development,

land and water management and forestry/energy. In Karatu, natural forest conservation

and indigenous cattle husbandry also covered more than 25% of the villages.

81



However, coverage in terms of numbers of villages gives a wrong impression of the pro-

portion of population actually reached. Several activities involved only a few farmers per

village. All in all, the total farm households covered in the various agricultural sector

activities do not exceed 5% of all farm households in Mbulu and even less in Karatu.

In environmental conservation, figures seem impressive at first sight: e.g. more than 60%

of farmers have been trained, although only 3% of the acreage under regular cultivation

has been affected. The proportion of teachers included in up-grading courses was also

high (85%), but no information is available on the effects of the training of either farmers

or teachers.

8.5 Impact on poverty

Obviously, one would like to know what all this output over a 15-year period has meant for

the well-being of the local population. Have standards of living been improved in

accordance with MDRDP’s main objective ?

Relevant information in this regard is very limited. The programme itself only started to

record data on impact in conjunction with the better monitoring of activities and output.

These are no more than impressions, based on limited observations, and are not related

to other factors that might have contributed to the observed changes in, for example,

agricultural yields and/or school attendance. Only the impact of the productive activities

of MDRDP has been more thoroughly assessed in a limited number of villages but often

without comparing project and non-project farmers. Based on the observed improve-

ments in soil and water conservation, improved land management, increased yields and

milk production as a result of better input use and crop and animal husbandry, an impact

study concluded that living standards in Mbulu district had improved as a result of

MDRDP (ETC, 2000)

The Mbulu District Development Plan for 2001 gives some indication of the effects of

social expenditure in the period 1998-2000, although without reasoning or analysis to

justify the claim that changes were the result of DRDP interventions. In the health sector

effects included an increase in ante-natal clinic attendance (from 70 to 80%), immunisa-

tion of infants (from 60 to 84%), family planning acceptance rate (from 1 to 2%), and a

decrease in the maternal mortality rate from (300 to 290 per 100,000) and in the inci-

dence of malaria and diarrhoea. In education, effects included improvements in pupil

performance in national examinations (from an average of 70 to 90 marks out of 150),

and an increase in the primary school enrolment rate (from 75 to 90%). However, various

sets of figures are in circulation but their reliability is not known.



82

The stakeholders involved in the programme show different perceptions of the role and

impact of MDRDP. A distinction can be made between the views and opinions of civil

service employees (Heads of Departments), elected Councillors, including the Mbulu

Council Chairman, village leaders and villagers (Chachage et al., 2001).

Government personnel

Senior government staff involved in the district largely share the view of donors. The

majority feel that the lack of financial means and expertise in the district necessitates

development aid. They support the need for “development engineering” through formal

planning and expert intervention. This category is inclined to assess positively the fruits

of interventions, attributing shortcomings mainly to annual delays in the release of donor

funds and to conservative views held by the farmers. To demonstrate their success, they

refer to the scale or number of activities that are conducted under the various projects.

Village leaders

Village leaders generally share the enthusiasm and optimism of government staff about

the role of development intervention. Their main concern is about the consistency and

equity of its distribution, because their credibility among their supporters depends on

such aid. They operate as brokers between the village and the outside world, and are often

elected because of their knowledge of government and donors. Their complaints involve

promises that were not kept, agreements that were not fulfilled, and terms that were

exceeded by government and donors alike. The following quote is relevant in this respect:

“ … a white man came here and talked about development … we agreed with him … But it has been a

long time since we last saw him. He has left us high in the air …” (Chachage et. al.,2001)

Councillors

Elected members of the District Council also lobby to gain as much available aid as

possible to benefit their constituents. They want to show tangible results of their efforts

to their supporters (schools, bridges, wells, clinics), and complain that too much money

is spent on meetings, seminars and training courses. Some councillors put the question

of donor intervention in a wider perspective They consider development intervention as a

tripartite relationship involving donors, the District Council and the target populations.

In the process each of these parties has a role to play. With regard to the donor, the

Council expressed concern about unilateral actions by the donor, namely, the extremely

strict requirements (unilateral) laid down for planning and programming, so that parti-

cular funds only become available a few months before the end of the budget year, and

83



about the sudden decision by the donor to restrict the role of the Dutch co-ordinator to a

purely advisory one. Some councillors articulated several shortcomings of the council,

including conservatism resulting from lack of exposure and of formal education. They

also mentioned the problems of the Council in taking over tasks because of acute staff

shortages, and the lack of experience and dedication on the part of available staff. In their

opinion, Mbulu still lacks the staff and organisation to support and manage a pro-

gramme of the present scale and intensity on its own, as is foreseen for the near future.

Rural population

The group for which MDRDP is intended, i.e. the rural population, seems to have different

expectations of development programmes than do the district authorities. Most ordinary

villagers do not distinguish between donor and government interventions, also because

all aid is channelled through government departments. When they do distinguish

between donor and government, they see government personnel as an obstacle to an

appropriate flow of development resources to the population. They mention intervention

by district officials in the distribution of development resources at the village level and

complain about the mishandling of food aid in times of drought and famine (Chachage

et al. 2001). The opinion of the population may be illustrated by the following quote: If

intervention comes from the top, it never reaches the target group. And the group is waiting but the

thing vanishes We get clouds instead of rain (video).

Three major concerns of the population have been distinguished (Chachage et al, 2001):

(i) The top-down planning process

There is a lack of transparency about the nature and scope of development projects and

the financial resources committed to these projects. Interviews with villagers show that

they rarely participate in development planning, also because meetings are held in

KiSwahili. Although many get the opportunity to express their opinion regarding develop-

ment priorities, it is unclear what is done with those opinions. All decisions are taken by

government officials who consider participation by villagers a compulsory but pointless

exercise.

This lack of transparency creates suspicions of corruption and distrust of leaders.

Respondents repeatedly narrated confirmed or alleged cases of the embezzlement of

funds or materials intended for village development. “If the donors want to help us, they

should give it (the aid) directly to the village, since then everybody knows how much money is

concerned … we must shorten (the lines of ) communication”.



84

(ii) Neglect of the productive sector

Villagers expressed growing concern about MRDP’s neglect of the productive sector and

limited attention for the improvement of agricultural and livestock services. They also

referred to soil erosion, and environmental degradation and complained that the District

Council/MDRDP hardly seems to address this problem. This perception does not

correspond with the attention and level of expenditure in MDRDP for environmental

conservation. The disparity is probably due to the dissatisfaction felt by the vast majority

of villagers regarding the performance of the district management in general, and the

departments involved in particular. In the final analysis, many rural people see the district

management team not only as negligent of their needs but also as colluding with the

emerging stream of business people in exploiting them.

(iii) Taxes

MDRDP has greatly emphasised the contribution to development activities to be made by

the population and district. The central and decentralised Governments of Tanzania are

making efforts to improve proceeds from taxation as part of its support to “good gover-

nance”, to increase the financial strength of the district and to make it less dependent on

donor aid. For the Netherlands support to the district is linked directly to the Council’s

investments in the respective areas, which in turn depends on council revenues and taxa-

tion. The population, although traditionally used to tributes, feels that the multitude of

taxes, dues and fines that are currently imposed, and the harsh way in which these are

collected, are out of all reasonable proportion. Taxes, which are not differentiated by

income, are a heavy burden, especially for poor farmers. They do not feel that they benefit

from the activities for which the development levy is intended, nor do the interventions

result in any production or income increase. To them, the result of the taxation is

increased poverty or, as some say, government exploitation. In 2002, the development

levy has been abolished.


The Mbulu DRDP is characterised by a very broad range of activities of a rather fragment-

ed nature, and shifts in focus over the fifteen years of its operation. These shifts from

support to production-oriented activities, to community services in the mid-1990s were

caused by a learning process within the programme, the priorities of politicians and

government staff, and changes in Dutch aid policies. They did not result from any long-

term strategy towards poverty alleviation or from a thorough monitoring and evaluation

that would measure progress towards achievement of the main objective of the

85



programme. Over time, the number of projects and activities per project increased and

resulted in a fragmented programme. There have been improvements in terms of better

soil and water conservation, improved land management, higher yields, more improved

varieties, and better dairy farming with higher milk output. The concrete effects of a par-

ticular intervention are not known. Firstly, the programme itself started only very recently

to record some information on impact in conjunction with the better monitoring of activi-

ties and output. These are still more or less impressions based on limited observations

and are not related to other factors that might have contributed to the observed changes.

Benefits have not been related to cost per intervention or per farmer; for many interven-

tions, however, these costs may be expected to be quite high in view of the low coverage

for several activities. Moreover, evaluations and especially the recent mid-term review

have added little to the understanding of impact, despite an elaborate system of self-

assessment that preceded the actual exercise. Second, other factors than DRDP-support-

ed interventions may have influenced the outcome. For example, it is not justifiable to

attribute to MDRDP a doubling of yield and income as a result of irrigation improve-

ments, if a substantial part of the acreage was already under irrigation before the project

started.

Employment and incomes do not seem to have improved during the last 10-15 years, nor

does food security. Coverage and outreach remained small and there has been no major

breakthrough in technology or production methods. Under the circumstances, few farm-

ers can afford to adopt modern inputs and few have been successful in re-introducing

cash crops. Even successful farmers continue to face serious constraints such as disease

and marketing deficiencies. Support has apparently been given in an environment where

production and incomes suffered from negative climatic and institutional arrangements.

National agricultural and livestock research institutes failed to come up with appropriate

improved technologies and field staff faced retrenchment. Crucial functions of a with-

drawing state were not replaced by a well-functioning market. In general, yields declined,

and existing levels of extension, input supply and marketing were negatively affected by

the decline of public services. Although primarily a national policy issue, the concrete

effects of this phenomenon at district level have hardly been addressed by DRDP.

In the social services sectors also, the programme has resulted in improvements in

infrastructure and staff training but little is known about how these affected the quality of

services. Nor is it known how sustainable the social improvements may prove to be in

terms of financial and institutional maintenance arrangements.



86

In the area of institutional changes there are no clear indications that local government

has been strengthened during the last 15 years as a result of DRDP, although it is at pre-

sent the major justification for the programme. The self-assessment of 2000 was even

somewhat negative in this respect, noting the downward accountability of councillors as

questionable, little participation in planning, and the fact that most departments did not

meet their targets in capacity building.

Concrete information is not available about the overall effect of DRDP on poverty reduc-

tion; in view of the low coverage for income-generating activities, the unknown effects of

improvements concerning social services, and the absence of targeting on the poor in

DRDP interventions, however, it is not likely to be positive. The assumption of a poverty

reduction effect from the strengthening of local government is still to be proven.

Moreover, the introduction of the matching principle between the volume of Dutch aid

and council investment funds intensified local tax collection and may have resulted in an

increase in poverty for certain segments of the population. Therefore, the conclusion of

the recent impact assessment that living standards in Mbulu District have improved as a

result of DRDP, remains doubtful.

Finally, it is somewhat ironic that DRDP seems to do exactly what it wanted to prevent in

the first place, i.e. to bring about social improvements without the necessary increases in

income with which to fund the provision of social services in the long run. In that respect,

the improvement in tax collection may prove to be a questionable success, as many

people complain about a higher tax burden without an improved income. The result

could be increased poverty for those concerned or, at least, a disincentive to improve

one’s own economic situation.

87





88

9 Economic change and localgovernance in Songea district

9.1 District profile

Songea is one of three districts in the Ruvuma Region. It covers an area of 33,925 square

kilometres, which is slightly more than half the total area of the region. According to the

1988 census, it had a population of 253,367, rising to 333,055 according to the 2002

census. Since July 2003 Songea District is divided into two new districts: Songea and

Namtumbo.

The main ethnic groups in Songea are the Ndendeule, the Ngoni and the Yao. The

Wandendeule were a stateless people with a strong egalitarian culture and ideology,

organised along several hundreds of matrilineal kinship groupings. They lived in the area

when, in the mid-19th century, two groups of Wangoni invaded the region from South

Africa. These had a strongly hierarchical and military organisation and established their

control over the Wandendeule and related peoples, who were incorporated in their states

under Ngoni chiefs. The Wayao crossed the Ruvuma river from Mozambique in the late

19th and early 20th century as hunters and traders. As former slave and ivory traders, they

had well-armed bands and established chiefdoms in a zone along the coast and the long-

distance caravan routes. Their different origin, history and culture can still be recognised

in present-day Songea. For example, religion and ethnicity are closely connected: the

majority of the Wayao and Wandendeule are Moslems, the Wangoni are mostly

Christians.

Geographically, the district shows considerable variety. It has three altitude zones:

a lowland area below 600 metres along the Ruvuma river basin (30% of the district), an

intermediate zone between 600 and 1200 metres (about 60%) and the highlands zone

between 1200 and 1800 metres (10%). Temperatures vary with altitude and average

between 20 and 25 degrees Centigrade in the daytime and 15-17 degrees at night.

The district experiences one long rainy season between November and May. Annual

rainfall averages between 800 and 1200 mm. but increases with altitude; hence, the

lowland is drier than the highland. Soils are poor in nutrients; almost two-thirds of the

area has soils with very low fertility, and less than 10% has medium to high fertility soils.

89


Economic change and local governance in Songea district



90

Map 3 Songea District

Road conditions in the district were poor, but have improved greatly over recent years.

About three-quarters of the road network is now passable all year round. In the rainy

season, the earthen roads are in poor condition, which is generally considered a serious

constraint on the marketing of agricultural produce, particularly in relation to the

district’s peripheral position. For example, although in 2002 the Government of Tanzania

opened the boundaries to neighbouring states for trade in food crops in response to the

famine in Southern Africa, this trade is still severely hampered by the poor communica-

tions network.

The economy of Songea District depends almost entirely on agriculture, which provides

employment to 95% of the population. The production of food and cash crops accounts

for more than 90% of the district’s Gross Domestic Product. Yet only about 10-20% of the

3.2 million hectares of arable land are currently under cultivation. The proportion differs

between one year and the next, the traditional farming system being shifting cultivation

with long fallow periods, and with cultivation mainly by hoe. The villagisation policy of

the late 1960s resulted in a more permanent agricultural system around relatively large

villages of approximately 300 households.

Smallholders cultivate an average of 3-4 ha. per household, dependent on available

labour. Women provide more than half the farm labour input. In the maize and tobacco-

growing zones, up to 45% of households use hired labour during peak periods. The main

food crops are maize and rice. Maize was introduced in the 1950s and strongly expanded

in the 1970s as a result of heavily subsidised fertiliser inputs. It is now the main food/cash

crop and is grown by all households. Tobacco is the second cash crop, cultivated by half

of all households. Paddy becomes increasingly important as a second grain crop, and

small quantities of cashew nuts, beans, simsim, and coffee are also produced. Livestock

production is insignificant compared to arable agriculture, but dairy farming has been

introduced and expanded since the late 1980s.

There are substantial differences in cultivated area, yield levels, degree of agricultural

commercialisation, and income, among farmers. Labour availability is the main determi-

nant for farm size and cropping pattern. Although income figures are not available, it is

evident that the highest agricultural incomes are found in the maize and tobacco-grow-

ing areas of Hanga and Udendeule divisions. Although with generally sandy soils that are

poor in nutrients, these areas are quite suitable for tobacco growing. The areas are

sparsely populated and show an unstable settlement pattern. The forests provide

91



abundant land for a ‘shifting-cultivation’ type of tobacco growing, which makes it easy to

avoid various pests and crop diseases and deterioration of the soil structure. The forests

also supply the firewood needed for tobacco curing.

9.2 Economic change

Subsistence agriculture characterised Songea’s rural economy up to the early 1930s,

when tobacco was introduced but this was only produced in small quantities in the 1930s

and 1940s. During the German colonial period and up to the early 1950s, labour migra-

tion was the main source of cash income and the means with which to pay taxes. Sisal

plantations elsewhere in the country were the principal destination for migrants; in the

late 1940s, no less than one-third of the adult male population was away at any time of

the year (Gulliver, 1955). As tobacco growing expanded in the 1950s, labour migration

was replaced by cash crop cultivation, especially after the British administration founded

the Songea Native Tobacco Board and established a seed farm in the area. At the same

time, moreover, the Ngoni-Matengo Co-operative Union (NGOMAT) started to operate a

tobacco–processing factory.

The Ndendeule in particular benefited from the new opportunities; their egalitarian social

structure and the new methods of collective action made them the keenest members of

the co-operatives. At the time of Independence, therefore, conditions for further agricul-

tural commercialisation and economic development in Songea were quite favourable.

The Government of Tanzania approved that tobacco should become the main cash crop

for Songea, and attempted to stimulate expansion by legal means. In 1965, it introduced

a by-law that compelled every household to grow a minimum of one acre of tobacco.

Although most villagers resented the law, tobacco production increased by some 13% per

annum between 1964 and 1971. Due to the compulsion, growth was caused by an expan-

sion in cultivated area rather than by yield increases induced by technological innovation.

Tobacco production fell from an average of 4000 tonnes in the early 1970s to less than

2500 tonnes in the early 1980s, and further to a mere 400 tonnes in 1985. Production was

affected unfavourably by the dissolution of the co-operatives, the decline in use of hired

labour, considered to be an exploitative relationship, and farmers’ preference for maize.

In spite of tobacco cultivation, Songea was one of the least developed parts of Tanzania

until the early 1970s. From then, the economic situation improved as a result of the pro-

motion of maize cultivation, stimulated by heavily-subsidised fertilisers made available

by donor grants. Maize prices improved over the years due to subsidies on inputs and an



92

assured market. Moreover, fertiliser use almost doubled yield levels between 1977 and

1986, i.e. from 1.4 tonnes to 2.5 tonnes per hectare. Songea became one of the main

maize producing areas for the domestic market.

By the end of the 1970s, however, the disadvantages of the system had become increa-

singly visible. There was an over-supply of fertiliser, tonnes of which lay unused in village

stores because villagers had withdrawn from cash crop production, and maize rotted in

Songea stadium due to the government’s inability to arrange transport. Simultaneously,

the country had to import maize to meet food shortages elsewhere. Not only did agricul-

tural production decline in Songea, consumer goods were also hardly available to the

rural population, which in turn discouraged farmers from growing crops for sale.

The economic reform policy became noticeable in rural Songea in 1981/82 when in

addition to village co-operatives also private traders were allowed to buy food crops.

Pan-territorial producer prices were then abandoned in 1982/83, and co-operative

societies in the form of regional co-operatives and as affiliates of CCM (the only political

party) were re-introduced in 1984. The Ruvuma Co-operative Union (RCU) started to buy

crops from farmers in 1985; it sold its maize to the National Milling Corporation (NMC),

and tobacco to the Tobacco Authority of Tanzania (TAT). By 1988/89, the marketing of all

food crops had been decontrolled and a competitive, multi-channelled marketing system,

including private traders, introduced.

The abandonment of the pan-territorial price system and subsidies on inputs was highly

unfavourable for peripheral areas, such as Songea with its infrastructure impediments

and high transport costs, although accessibility improved in the late 1980s with the con-

struction of the Njombe-Songea road. The discontinuation of subsidies in 1986 reduced

the use of inputs, with an unfavourably effect on yield levels and quality of produce. The

NMC refused to purchase RCU maize or bought only part of it, because of its alleged low

quality. RCU incurred enormous losses and, by the end of the 1980s, was in serious debt

to the banks and government. It stopped purchasing food crops in 1991 and concentrated

instead on tobacco.

Despite the abolition of subsidies and continuous problems in marketing, official statis-

tics state a maize production fluctuating between 150,000 and 200,000 tons between

1988 and 1998 with a sixfold increase in value during the first half of the 1990s. A recent

report mentions a fall in yields to 0.7 tonnes per ha on an average 1.1 ha per household.

However, these figures do not tally with output levels (see Table 9.1): either households

93



cultivate a larger area or yield levels did not decline so dramatically. Even now it seems

that fertiliser use is considerable: survey results indicate its use by 80% of farm house-

holds (ETC, 2001a).

The decline in real producer prices for maize stimulated farmers to focus on tobacco culti-

vation again. Tobacco production increased strongly, although figures differ substantially

among the various sources. The most conservative indicate a volume increasing from

some 400 tonnes in 1985/86 to 3500 tonnes in 1989/90 and further to almost 7200 tonnes

in 1999/2000.

Marketing continued to create problems for farmers. Private traders began to operate

legally in tobacco purchasing in 1996. They provided inputs on credit, deducting the cost

when purchasing the crop. There was considerable distrust between farmers, private

buyers and leaders of the co-operatives, however, especially about the grading practices

used by private buyers. Nevertheless, the initial effect of increased competition between

private traders and the Songea Agricultural Marketing Co-operative Union (SAMCU),

operating at the level of Songea District, seems to have been beneficial to farmers.

SAMCU paid a slightly higher price than the average indicative one set by the Tobacco

Council. Politicians intervened when the district administration claimed that private

buyers did not pay the tobacco cess to the district coffers and court cases followed to

press the buyers to pay, and in 2000 SAMCU was granted sole rights to buy tobacco.

SAMCU did not distribute inputs, however, was criticised by growers for incorrectly

grading the crop and handling the funds, proved unable to find buyers for the tobacco

and, in the end, did not pay or only partly paid the tobacco growers. Subsequently, many

farmers stopped growing tobacco or reduced their area substantially. In 2002 two private

buyers were allowed to buy tobacco again, receiving a monopoly position in the area

allocated to them and reaching agreement with the district about the payment of cess.

Farmers still complain about the grading of the crop, however.

Paddy production has increased during the last decade as a result of improvements in

irrigation and crop husbandry techniques. It now covers some 10,000 to 16,000 ha, on

which 20-25,000 tonnes of paddy are produced annually.

Table 9.1 presents production figures for the two principal marketed crops.



94

If measured by the value of marketed output, the incidence of mechanisation and the

level of input use, Songea has a relatively low level of agricultural commercialisation,

especially if compared with potential and available land resources. A comparison of

actual and potential yield levels, as established through research trials, shows ample

potential for output increases. Gross margin calculations based on actual prices received

by farmers indicate interesting prospects for tobacco, irrigated paddy, coffee and ground-

nuts. Conversely, maize proves to be the least remunerative crop. One of the main factors

responsible for low gross margins is low productivity (ETC, 2001a).

9.3 Social change

Economic change also affected the social system in Songea District. Labour migration to

the sisal estates and the cultivation of cash crops for the market expanded the monetary

economy. This increased socio-economic differentiation and changed the relative

position of the three main ethnic groups, chiefly in favour of the Ndendeule. In addition,

missionary organisations and the state started to provide education and health services.

95



Table 9.1 Production and value of main crops, Songea District 1988–2000*

Maize Tobacco Total Value

Year Volume Value Volume Value

1988/89 178.7 1.07 3.5 - -

1989/90 147.2 1.32 3.2 - -

1990/91 134.9 1.62 5.3 - -

1991/92 199.6 5.99 3.4 1.56 7.55

1992/93 195.8 5.87 3.5 1.71 7.58

1993/94 206.0 8.24 4.3 2.14 10.38

1994/95 154.8 6.97 4.8 2.64 9.61

1995/96 156.2 7.03 8.0 4.40 11.43

1996/97 79.6 4.62 9.2 4.60 9.22

1997/98 176.2 12.33 4.6 2.5 14.3

1998/99 - - 7.4 4.2 -

1999/00 - - 7.2 4.4 -

* Volume in 000 tonnes, value in billion Tshs, - = no data

Source: District Development Plan, 2001

Education started in Songea with the coming of the missions, mainly the Roman Catholic

Church, which became established in Peramiho in 1898. Even now, Peramiho schools are

well-known in Tanzania for good quality education. The number of mission schools and

their high standards had a positive influence on the education level in Songea. The dis-

trict has always had a high percentage of school-attending children. After Independence

and especially in the early 1970s, schools increased in number, but accurate data on the

magnitude of that increase are lacking.

Since the 1980s, however, there has been a decrease in enrolment from about 90% to 70%

in 1996, subsequently increasing again to 81% in 2000/01. Enrolment rates in primary

education fluctuated in relation to population growth, availability of classrooms and the

accessibility of schools in terms of distance and the level of school fees. During the last

decade, more girls than boys have entered education; boys, moreover, show a higher

drop-out rate. At present, enrolments and the quality of education are both affected by a

shortage of qualified teachers, classrooms and school materials, particularly in the more

isolated villages.

Compared to national standards, the number of health facilities in Songea is rather

favourable. Since the 1970s the number of dispensaries has more than doubled (from 26

to 64), and the number of health centres has increased (from 1 to 7). About half of all

dispensaries and health centres are private (mission). There are two hospitals, located

close to one another. This complicates access to hospital care for rural areas, especially

during the rainy season. Although facilities have increased and are more evenly distri-

buted throughout the district, the quality of health care has fallen since the early 1980s.

The health referral system is reported to have collapsed and village dispensaries face

shortages of staff and medicines.

Since Independence the construction of water supply facilities has increased, facilitated

by donor aid and self-help contributions of the local population. Maintenance has often

been a serious problem, especially since the mid-1980s when the Tanzanian economy

declined and public expenditure had to be reduced. Consequently, clean and safe water

coverage decreased from 54% of the population in 1988 to 47% in 1996. Subsequently, it

improved again with the support of donors; at present it is claimed to be in the order of

70%.



96

9.4 Local governance

Before the German occupation, Ngoni invaders controlled most of the area of what is now

Songea District. This was facilitated by their hierarchical socio-political structure and

their military capabilities and strength. The system of administration adopted during the

German colonial period affected the position of ethnic groups in several ways. Firstly, it

reduced the powers of the Ngoni vis-à-vis other ethnic groups. The Germans prohibited

slave raids and the slave trade, i.e. the main source of income for the Ngoni and Yao.

The Ngoni reacted in two ways: some tried to co-operate with the Germans and sent their

children to mission schools, others refused to adjust to the rules and regulations of

colonial administration and ultimately revolted against the Germans, in what became

known as the Maji Maji rebellion (1905/06). The crushing of the rebellion meant the end

of the military power of the Ngoni and resulted in the migration of large groups. Famine,

malnourishment and disease depopulated parts of the area. The Ngoni had converted to

Christianity at an early stage and many trained tribal members were later accepted by the

British in their administration.

The British administration introduced Native Authorities in 1926 under the principle of

Indirect Rule. Songea became then a separate district with four native authorities. The

policy of Indirect Rule favoured ethnic groups with a hierarchical socio-political structure,

such as the Ngoni and Yao. As late as 1953, also a Ndendeule native authority was

established. Decentralisation policies under British administration gradually transferred

powers to the Ndendeule, numerically the largest ethnic group in Songea, at the lower

administrative levels, particularly the division and village. During the struggle for

Independence, Ngoni and Yao chiefs became more loyal to the British administration,

whereas the demand for a fully representative District Council arose among the

Ndendeule, whose political position strengthened when chiefs were abolished by the

independent Government of Tanzania. The Ndendeule had already strengthened their

economic position during the 1950s by taking a leading role in tobacco cultivation and

marketing.

The situation again changed with the re-centralisation of political power after the adop-

tion of socialist policies, particularly after 1972 when local government and its elected

officials were abolished and replaced by a de-concentrated government bureaucracy from

the centre down to the village level. From 1973/74 onwards, Songea faced a massive

villagisation programme, because of the vast area of uncultivated land, scattered popula-

tion, the low level of social services and poor communication. Therefore, most of the

population were forcefully moved into new villages. Villagisation had different effects on

97



the various ethnic groups. It strongly affected the Ndendeule, who used to live in small

and dispersed groups, but brought little change to the Ngoni and Yao, who already lived

in villages. Under the new regulations, villages were to have elected councils, with several

committees to manage, regulate and co-ordinate development affairs and to mobilise

villagers for social and economic development. The village was transformed into a

co-operative society, a production unit rather than a site of governance.

During this period, the co-operative system was also re-organised, implying a reduction

of farmers’ influence. The centralisation tendencies already became apparent during the

late 1960s with the dissolution of the Ruvuma Development Association (RDA), a group of

17 communal villages, that had been established in 1963 and turned into a co-operative

body in 1965. Villages were successful in their development efforts in terms of improving

food security, establishing village industries and improving social services, such as

education, health and drinking water facilities. RDA activities had become a threat to the

co-operatives and were weakening the position of politicians. Thus, in 1969, the Minister

for Rural Development and Regional Administration announced that RDA was dissolved;

its assets were confiscated with the aid of the police.

The decentralisation of local administration was re-introduced in Songea in the early

1980s and co-operatives were re-established. Local government elections were integrated

into the one-party structure, and co-operatives also operated within the party framework.

Radical changes to this structure were to take place in 1992 with the abolition of the one-

party system and the introduction of multi-party democracy. Formally, government and

party structures were separated at the village, district and regional level. In practice, in

Songea, 99% of village governments continued to be controlled effectively by CCM.

According to the Election Monitoring Committee, it was not uncommon that village

governments worked together with CCM to harass those who joined other parties.

The situation does not seem to have changed up to the present day. In most villages,

CCM is still firmly in control and dominates leadership and other issues (Chachage et.

al., 2001).

The main elected body, the District Council, is composed of 38 councillors. Out of these

26 represent wards (one is a woman) and two are Members of Parliament. In addition,

there are 10 nominated female councillors. However, women tend to play a passive role in

the various organs.



98

According to recent reports, the Council does not have a clear view on the development of

the district or on its own role in the development process. Councillors lack facilities, such

as an office and adequate means of transport to visit their electorates and do not repre-

sent the people in the way envisaged. The District Management Team, although strongly

committed to change, is plagued by frequent transfers, unclear job descriptions, the lack

of incentives to motivate staff, and promotions that are not based on performance. At the

village level, cases of poor leadership, malpractice and corruption have been observed,

and it was reported that problems were not solved because villagers fear their leaders.

Regarding financial management, an audit by Price Waterhouse Coopers concluded that

“the Finance Department appears to have no strategies on how to manage the finances it

collects or how to better collect and co-ordinate the collection of revenue” (SDC, 1998).

Village studies show that weak financial management at the village level is a serious

problem: it discourages the population from participating in communal activities and

hampers the development process. Respondents report embezzlement of funds, and the

use of people’s contributions by village leaders and civil servants for allowances,

festivities and other types of private consumption (Chachage et. al., 2001).

9.5 Perceptions and poverty

The economic history of Songea District has been determined by the agricultural sector,

which was and still is the dominant sector of the rural economy. In the first half of the

20th century, agriculture was limited to self-sufficient food production and cash incomes

originated from the labour migration of adult males to sisal and other plantations

elsewhere in the country.

The second half of the 20th century shows the replacement of labour migration by the

cultivation of tobacco and maize as main sources of household income. During this peri-

od, there were sharp fluctuations in the levels of agricultural production and monetary

incomes of farmers. In the years preceding Independence, there was a strong expansion

of tobacco growing which continued up to the early 1970s. Incomes of tobacco growers

rose and their living conditions improved. Yet, in those days the area was considered as

one of the least developed in Tanzania due to its remoteness, inaccessibility and low

population density. From the early 1970s onwards maize production rose dramatically,

the economic situation improved and incomes increased, particularly as farmers profited

from subsidised fertiliser inputs. Economic reforms, the abolition of subsidies on fertilis-

ers in particular, affected farmers’ incomes unfavourably from the mid-1980s onwards.

If the aggregate value of major crops is used as a main indicator, average per capita

99



incomes fluctuated from about US$ 50 in 1995 to US$ 90 in 1999/2000. Other sources

refer to estimated incomes of US$ 100 and US$ 250 per capita. These averages conceal

income differences between various parts of the district, and among farmers, i.e. espe-

cially those with or without hired labour. Moreover, price fluctuations throughout the year

may affect farmers differently. Most low-income farmers have to sell their farm produce

immediately after harvesting (when prices are low), because of a need for cash and lack of

adequate storage.

Although reliable data on incomes are lacking, it is obvious that per capita incomes are

considerably below the poverty line and that incomes have decreased since the 1980s.

The improved housing conditions for a majority of households, mentioned in the village

profiles, are the effect of higher incomes in the past, especially in the 1970s

(Chachage, 2001; ETC, 2001a).

There is general agreement among the various stakeholders in the district that poverty

has been increasing in recent years. In the villagers’ perception, poverty manifests itself

in terms of low income or no income at all, the inability to have good modern houses, the

inability to feed and clothe members of the household, to send people to school or to

afford school fees and medical services. According to one of the farmers:

Lack of seriousness in the buying of tobacco has made people stop cultivating the crop. There is no

hope. Most of the villagers are now growing maize, but maize is no longer a cash crop since there

is a surplus at the local level and marketing to other areas of Tanzania is too costly (Chachage

et. al., 2001).

A recent community assessment of agricultural production, income and livelihood

presented a similar picture. In 1998/99 half the respondents assessed their situation as

bad, and this increased to three-quarters in 2000 (ETC, 2001a).

The reasons given for the recent increase in poverty differ. At the regional and district

levels, authorities attribute further impoverishment to the under-development of

industrial and commercial sectors. Crop yields are low because of the inadequate supply

and use of inputs and low technology of production. The business community is of the

opinion that investment in non-agricultural activities is not possible because of insuffi-

cient power supply, whereas the power company states that there are no activities to justi-

fy investments to increase the capacity. Consequently, tobacco buyers find it more conve-

nient to process tobacco elsewhere. At another level, district authorities and the District

Rural Development Project (DRDP) regard the question of poverty as being caused by the

lack of alternative cash crops to tobacco, given that the market for tobacco is not secure.



100

Villagers claim that if any real development and improvement of their lives occurred at

all, it was in the second half of the 1970s. That period was marked by good leadership,

and they were able to sell all their produce. Although co-operatives delayed payments,

they were assured of being paid eventually. Moreover, social services were expanded and

improved in the 1970s, and therefore the people perceived that period as one of decreas-

ing poverty. Furthermore, re-centralisation was seen at the village level as positive:

villagers claim that during this initial period villages were able to develop, “because

leaders used to listen and deal with people’s problems” (Chachage et.al., 2001).

In the perception of the village population, present poverty is mainly caused by the lack of

access to inputs and the absence of a reliable market for whatever they produce, since

there are no organisations to look after their interests. Moreover, they feel that district

and regional governments are apathetic towards addressing their marketing problems

seriously. Their criticisms about the quality of present leadership include those who have

been elected, such as Members of Parliament and Councillors who, in their view, behave

like government officers. This is illustrated by the following quote from an interview in

one of the villages:

The honeymoon with the MP ended with our having elected him. He says he was not elected for

the purpose of solving minor problems- those at the village level…. He deals with national issues.

He says we are making a lot of noise to him, wowowowo. But we are adults. He better tells us,

who is our MP ?… All that we told him was, a tobacco farmer is treated unfairly when marketing

the crop. It is not fair that a farmer is not allowed to witness and instead is represented, when

tobacco is being weighed and graded. What kind of trade is this, when one is not allowed to keep

an eye on his property?

These opinions are confirmed by the ‘green sector’ study which recorded the complaints

of farmers about their perceived ill-treatment in tobacco marketing, the ineffective or lack

of action by district authorities, and the numerous taxes levied on their crops. According

to that study, farmers reacted with a survival strategy, including defaulting on inputs,

rather than by adopting better farming methods (ETC, 2001a).


Songea District has a rather favourable resource position in terms of land resources, but

suffers from a peripheral location vis-á-vis the main population centres in the country and

from a poor communications network. The district’s economy is dominated by the agri-

cultural sector, which provides employment for 95% of the population and represents

101



90% of production. The growing population has gradually expanded the area under

crops, but there is still ample land available.

Rural economic change resulted primarily from agricultural commercialisation which

started during the colonial period but was vigorously pursued by government after inde-

pendence. The principal instruments of government policy up to the mid-1980s were legal

regulations regarding minimum areas for crops, subsidies on inputs, a pan-territorial

pricing system that favoured peripheral areas such as Songea, and heavy state interven-

tion in input supply and marketing. This policy resulted in income increases and

decreased poverty in Songea, but was disastrous for the country’s economy and therefore

non-sustainable.

Economic reforms introduced after the mid-1980s brought the termination of subsidies

and the pan-territorial pricing system, and a decrease in real producer prices. The conse-

quence of these changes was a reduced use of inputs, lower yield levels and lower real net

farm incomes. In addition, the entry of private traders did not lead to greater competition

and higher efficiency of the marketing system. When marketing problems for tobacco

intensified at the end of the century, farmers lost interest and reduced their cropping

area. In general, economic reforms have not given rise to the expected increase in

agricultural production and farm incomes.

Social change involved a shift in power among the main ethnic groups, increased social

differentiation in relation to expansion of the monetary economy and the creation of new

economic opportunities, the spread of new religions (Christianity and Islam), and the

expansion of social services such as education and health care systems. A sharp expan-

sion of these facilities occurred during the 1970s and early 1980s, but their magnitude

and quality declined after the mid-1980s. During the 1970s, moreover, large groups of

dispersed households were pressed to start living in villages in order to facilitate the

supply of these services.

After Independence the population of Songea experienced an improvement of living

conditions due to higher monetary incomes, the construction of better houses, and the

expansion of social services. There was thus an obvious reduction of poverty, although

improvements were concentrated in selected parts of the district, more especially Hanga

and Udendeule divisions. After the mid-1980s and the introduction of economic reforms,

the output of main crops continued to increase but, because of higher costs and ineffi-

ciencies in marketing due to lack of competition, net farm income stagnated or even



102

decreased. Although average incomes in the region, and in all probability also in the

better-off parts of Songea, are above the national average, people experienced this period

of reform as one of increasing poverty.

There is a sharp difference of opinion about the main causes of the present crisis and the

appropriate strategy to improve the situation between the authorities on the one hand,

and the ordinary people on the other. The authorities emphasise the need to diversify the

rural economy through the expansion of non-agricultural activities and the introduction

of alternative crops. The local people consider the deficiencies in agro-support services to

be their main problem, especially those in the marketing system. Also, they stress the

need for substantial improvement in the quality of leadership at the local and district

levels.

103




Economic change and local governance in songea district

104

10 District rural developmentprogramme in Songea

10.1 Introduction

The Songea Integrated Rural Development Programme, later named Songea

Development Action (SODA), is a joint programme between the Songea District Council

and the Netherlands Development Organisation (SNV). The contents and set-up of the

district programme in Songea were formulated in 1991 on the basis of a fairly general

initial survey in the district. The programme officially started with the signing of an

Administrative Agreement by representatives of the District Council, SNV and the Office

of the Prime Minister in October 1993. The agreement covered the period 1993-1997, for

the first phase of the programme. The year 1998 was a preparatory year for the second

phase, which started in 1999 and is now ending in 2003.

10.2 Objectives and strategy

The general objective of the SODA programme for the first phase was formulated as fol-

lows: to support and promote a community based development process in geographical

areas of intervention in Songea Rural District, which will lead to sustainable improvement

of the living conditions of the rural population.

The specific objectives included (i) strengthening of government planning and imple-

mentation, (ii) creation and strengthening of people’s organisations at village level, (iii)

improvement of the agricultural production system to ensure food security and increase

cash income, and (iv) implementation of a health care and nutrition programme.

For the second phase the overall aim was “to structurally improve the well-being of men

and women by enhancing food security, rural income and improving social services in a

sustainable, efficient and equal way.’’ This general aim was worked out into three more

specific objectives, phrased as follows:

1. To further develop the participatory planning and management approach of commu-

nity-based projects, and to embed it successfully into the regular council planning

cycles in order to increase local ownership, stimulate local institutional development,

enhance project sustainability and enlarge the participation of women in planning

and management.

105


District rural development programme in Songea

2. To enhance the management, planning, implementation and monitoring capacity of

the District Council in order to fully hand over all project responsibilities to this

council.

3. To design and implement strategies for (i) creating an enabling environment for the

emergence of NGOs, CBOs and the private sector at district and local level; (ii)

improving of household income, food security and living conditions of the district

population; and (iii) strengthening the relative position of women in the develop-

ment process.

SODA adopted a strategy leading to short term tangible results. This meant the imple-

mentation of activities on the basis of a rapid rural appraisal, and a focus on local

communities and people’s participation in problem identification, priority setting for

improvements, development of action plans and implementation of activities in accor-

dance with these plans. Simultaneously, further studies were undertaken to identify

structural development constraints and potentials. To this end, the programme started in

39 villages in three of the seven divisions of the district. Divisions were sub-divided into

agro-ecological zones for which profiles were made, specifying resources, population and

farming system characteristics and information on basic services and gender. The profiles

also indicated the viability of interventions in the fields of agriculture, agro-forestry,

economic activities, accessibility and basic needs, compared to the priorities identified

through the local communities.

The programme had an explicit focus on poverty alleviation. According to the formulation

report, the determining factor for successful impact in terms of poverty reduction was

“the building of capacity of the target group to tackle in a concerted way the problems

that keep them in a situation of poverty’’. The main instrument through which to achieve

this was village level planning, which started with facilitating the village population in

analysing their problems and felt needs, followed by setting their priorities and develop-

ing action plans to solve the problems. The villages requested financial support from

SODA for plan implementation via the district planning officer. District departments were

usually involved in implementation.

In the second phase, the coverage of the programme was expanded to include the whole

district, while emphasis shifted from strengthening the local communities to enhancing

the capabilities of the district administration. This meant the incorporation of village and

ward plans into district planning, more attention for the training of local government



106

personnel, and increased implementation through sector departments at district level.

Moreover, project funding was replaced by direct funding of the district administration

for the implementation of the district plan, and a separate contract for SNV’s advisory

services.

10.3 Programme organisation and finances

Organisationally, the programme operates through Songea District Council. The District

Executive Director (DED) is in charge of a team of Heads of Department, responsible for

drafting a district development plan, which has to be approved by the Council prior to

submission to the Ministry of Regional Administration and Local Government. After

approval, the various departments implement the plan. In 2001 Songea District Council

had a staff of 2341 of whom 1654, or two-thirds were employed in education, mostly

primary school teachers.

During the first phase, the programme had a parallel structure with separate project

funding and technical assistance through the deployment of expatriate personnel. At the

district level SNV was represented by a programme co-ordinator. The Dutch co-ordinator

played a key role in shaping the programme, i.e. in the design of the strategy, in the

selection and implementation of activities and in the approval of budgetary allocations.

The DED and SNV co-ordinator both signed for expenditure. Apart from the co-ordinator,

SNV had a staff of five (two area co-ordinators, a specialist on women and development, a

farming expert, and a part-time project administrator).The various advisors were attached

to relevant departments of the district council.

The organisational structure changed during the second phase. The programme became

embedded wholly into the regular administrative structure of the district, technical assis-

tance personnel started to operate more as advisors and facilitators and were reduced to

one person in 2002. Most responsibilities were handed over to council staff, who made

the final decisions about financial allocations on the basis of proposals of the depart-

ments. The SNV co-ordinator attends council meetings for questions and advice.

Although SODA formally joined the DRDP programme supported by the Netherlands in

1991, actual implementation started in 1994, some three years after its formulation.

Project expenditure for the period 1994-1996 totalled US$ 2.20 million, including 0.75

million for SNV staff. Of the US$ 2.20 million of project expenditure, about one-third was

spent on administration and office management. During the next three years (1997-1999)

project expenditure increased slightly to US$ 2.5 million, while the proportion for admin-

107



istrative overheads gradually fell to 15%. This means that the share of development

investments increased from 65% in the initial years of the programme to 85% in 1999 and

further to almost 90% in 2001.

Expenditure for the whole programme period (1994-present) totalled an estimated US $

6.3 million, including expenditure on SNV staff input.

The distribution of expenditure over the various sectors changed in the course of the pro-

gramme. During the initial period, i.e. 1994-1996, no less than 42% went to the rehabili-

tation of feeder roads and bridges, followed by 21% for the improvement of drinking

water supplies, 10% for agriculture and natural resources and 8% for education. Together

with the 5% for health, combined expenditure on the social sectors represented about

one-third of the total. This was maintained during subsequent years, but there was a

drastic reduction in expenditure on roads and bridges and fairly sharp increases on agri-

culture and livestock, and on capacity building in the district administration. The latter

more than doubled in the period 2000-2002 and became the most important cluster of

activities. Expenditure on agriculture/livestock and land use was maintained at the same

level, whereas those for the social sectors and road infrastructure further decreased. This

shift from social services towards strengthening the district administration formed the

main deviation from the 2001/2003 budget.



108

Table 10.1 Total expenditure SODA/DRP 1994–2002 (US$ 000)

Year Development expenditure* Total expenditure Development expenditure

as % of total

1994-96 1445.0 2200.0 65.6

1997 684.9 878.8 77.9

1998 543.1 665.6 81.2

1999 771.8 909.4 84.9

2000 488.9 568.0 86.0

2001 493.5 550.9 89.6

2002 488.1 558.3 87.4

Total 4,915.3 6,331.0 77.6

* includes capacity building

Source: SNV records

The trend for sectors is aggregated in table 10.3 for the four main categories of expendi-

ture.

The sector distribution and trends of DRDP expenditure do not provide a complete picture

of development expenditure in Songea. Firstly, other donors support projects in the dis-

trict, for example for drinking water improvements and dairy farming propagation.

Second, sector support by donors, including the Netherlands, provided for national level

programmes, may become available to the district. For Songea this applies to road con-

109



Table 10.2 SODA/DRDP - Sector distribution of development expenditure 1994–2002 (%)

Sector 1994/96 1997/99 2000/02 Budget 2001/03

1. Capacity building /

Strengthening district administration 2.6 10.9 25.3 6.0

2. Roads / bridges 42.3 20.1 15.7 12.0

3. Agriculture / land use 9.7 21.9 23.6 29.0

4. Education 7.8 12.1 9.8 12.0

5. Health 4.9 7.8 5.5 9.0

6. Water 21.1 11.0 9.3 12.0

7. Community development – * 8.7 4.7 –

8. Other 11.6 7.5 6.1 10.0

Total 100.0 100.0 100.0 100.0

* Not separately mentioned

Source: SNV records

Table 10.3 Main categories of development expenditure SODA/DRDP 1994-2002 (%)

1994/96 1997/99 2000/02

1. Basic needs / community services 34 31 25

2. Infrastructure / communications 42 20 16

3. Income generation / economic base 16 33 33

4. Capacity building / district administration 3 11 25

struction through the road rehabilitation fund and to health and education. Priorities for

sector funding at national level may differ from district priorities, however; at present,

district plans do not form part of the framework for national level planning and funding

in Tanzania. The planning system at district level is complicated further by planning

activities under the Local Government Reform Programme. The District Planning

Department, therefore, has to take account of several planning frames, each with its own

time schedule, methodology and set of rules and regulations (ETC, 2001a).

DRDP progress reports indicate substantial differences among projects in budget realisa-

tion, but for most years during the second phase and for most activities 80 to 90% of the

SODA/DRDP budget was spent. The contribution by the Songea District Council has been

substantially lower than arranged in the agreement between it and the donor. Initially,

SNV support towards strengthening the council was based on a 1:3 matching principle,

lowered for SDC gradually to 1:5, or from 33% to 20%. Even this lower target has never

been realised, however: for example, SDC contributed only 12% in 2002. The low contri-

butions by SDC were largely due to deficiencies in revenue collection, more especially a

long lasting dispute between the council, and the agricultural marketing co-operative

(SAMCU) about the transfer of produce cess to the council’s coffers. In 2000 SNV, conse-

quently withheld outstanding funds for investments, including those for the rehabilita-

tion of post-primary schools, the improvement of traditional wells, and the propagation

of new crops.


The distribution of funding provides some indication of the relative importance of sec-

tors. Support to sectors comprised clusters of activities or projects, however, which dif-

fered strongly in magnitude and complexity. In 1998 the programme comprised no less

than 72 activities under 16 projects, grouped into 5 clusters. For detailed information, see

Annex 3, table 3.4.

The most complex clusters were those of agriculture and natural resources, and the

organisational strengthening of the district administration. The former included, among

other things, the propagation of new crops (coffee, coconut and cashew), the introduc-

tion of animal draught power and irrigation techniques, the improvement of agricultural

extension services, the introduction of improved animal husbandry (dairy farming,

piggery and chicken farming), the propagation of agro-forestry and timber production,

the revitalisation of co-operatives including the establishment of Saving and Credit

Co-operative Organisations (SACCOs), and small-scale enterprise promotion.



110

Strengthening the district administration comprised, among other things, such diverse

projects as improving district planning skills (5 activities), strengthening the revenue

collecting system (14 activities), the budgeting and income administration (3 activities),

policies and procedures (5 activities), and the establishment of a district development

monitoring system. Also, the drafting of a District Development Plan for 2000-2002 was a

separate project in this cluster.

The village planning exercise covered all 128 villages in Songea District. According to an

internal evaluation report of 1997, all stakeholders agreed that this planning exercise

proved to be an effective tool for participatory planning and contributed to empowerment

of the people because of the short planning cycle, its capacity building elements, its

transparent procedures and its adherence to village-level priorities. Later reports state,

however, that participation in planning was still low and needed to be streamlined “in

accordance with existing democratic institutions”. And implementation of village plans

was hampered by shortage of funds and the lack of capabilities among government staff.

Finally, an assessment of the cost-effectiveness of the exercise pointed towards a more

pronounced role for ward level planning. Subsequently, village level planning was termi-

nated and replaced by the preparation of ward level plans. By 2001 all 26 wards were

covered by Ward Development Plans, which were integrated in 2002 into the District Plan

(SNV, 2000).

Improvement of the physical infrastructure concentrated on the construction of bridges,

spot improvements of roads and the setting-up of maintenance groups. By 2002 three-

quarters of council roads were passable throughout the year against one-quarter in 1994.

In the support for the agricultural sector, the emphasis was on the introduction of new

crops, new types of animal husbandry, improved farming and animal husbandry tech-

niques, and strengthening of the extension service. Relevant for sustainable agricultural

development were also the land use planning activities. Land use plans were prepared for

some 20-25% of all villages, which resulted in the re-afforestation of erosion-prone areas

and a reduction of boundary conflicts between villages.

The acreage covered and number of farmers involved differed strongly for the various

projects; for example, between 11,000 farmers for cashew nut tree planting to 90 farmers

for improved paddy, but such figures do not always provide a realistic picture of the

importance and impact of the projects.

111



For the propagation of cashew nuts, seed was distributed to numerous farmers in one

division, but the acreage that was planted and the number of plants that survived are not

accurately known. Seed distribution was influenced by the relatively high prices for

cashew nuts in 1998/99 and by strong support of local politicians because of the parlia-

mentary elections of 2000. Total output did not exceed 50 tonnes in 2002; this is expected

to increase slowly together with the expansion of buying points by private traders. So far,

the cashew nut project had little impact.

In 2000 DRDP started a project to improve rice production through better irrigation tech-

niques, the introduction of new varieties and improved planning methods. In total 90

farmers in 19 villages were covered. Activities included the demarcation of areas suitable

for irrigation, the construction or rehabilitation of small dams and channels, a survey of

farm plots, and the preparation of nurseries and demonstration plots. In addition, in

2002 government started to support improvements in paddy cultivation, which were

rapidly adopted by an estimated 5000 farmers in the area with advice by the extension

service. Reportedly, yields have increased threefold from 8 to 25 bags per acre/harvest and

irrigation now allows for two harvests per annum. Paddy is marketed through local chan-

nels, and prices have been influenced favourably by recent food shortages in Malawi and

by the liberalisation of trade to neighbouring countries. In sum, improvements in paddy

cultivation had considerable impact on farmers’ incomes.

In the social sectors, improvements of health facilities, primary education and drinking

water supply were supported. Community health workers were trained in all villages of

the district, and a number of dispensaries and rural health centres (18), and staff houses

(14), were constructed or rehabilitated. Every ward now has at least one dispensary. In

primary education, efforts until 1997 were concentrated in the most remote and neglected

division of the district. From 1998 onwards the whole district was covered. Activities

included the construction or rehabilitation of schools or individual classrooms (some

200), latrines and teachers’ houses, the up-grading of teachers’ capabilities, and the sup-

ply of teaching materials. In addition, a school inspection system was established and

almost all schools in the district were inspected, and school committees were set-up and

trained. In the drinking water sector, emphasis was on the rehabilitation of traditional

wells (over 400 serving some 20,000 people), but hand pumps were also provided to

about 40 villages (serving another 18,000 people), i.e. to about 8% of the district popula-

tion. At present, some 70% of the population of rural Songea is provided with safe water,

largely due to the combined efforts of a Danish project, that has recently been termina-



112

ted, and DRDP. DRDP estimated that about 60% of all achievements can be attributed to

Danish support and the remaining 40% to DRDP.

The broad range of the cluster of projects aimed at strengthening government adminis-

tration reflects the shift to good governance characteristic of Dutch support to DRDPs in

Tanzania. In the initial phase, SODA placed most emphasis on the village level, although

some training of district staff was included in the programme and the district administra-

tion was involved in plan implementation. Several evaluation reports nevertheless stated

that more capacity building at district level was needed and that activities lacked a

coherent council capacity building plan.

Since the late 1990s strengthening the district administration has been the principal

objective. Activities have focused on improving planning capacity and financial manage-

ment. This included such diverse projects as setting-up a gender task force, overhauling

the carpool system, training councillors and staff in participatory planning, improving

the revenue collecting system and various types of computer training for regular and

financial administration. The district council was also supported in its implementation of

the Local Government Reform Programme, for which Songea was a pilot district.

Although it is difficult to measure the impact of these activities, an indication of the

improved performance in financial management is the higher proportion of actual rev-

enue collection against budget estimates. By sharply reducing budget estimates and by a

moderate increase in the efficiency of revenue collection, performance improved from

51% in 2000 to 92% in 2002.

The programme introduced some innovative approaches in agricultural credit, health

care and education. With regard to the former, the district signed contracts with village

governments whereby the latter agreed to act as administrator and guarantor for credits

for seedlings and other inputs provided to individual farmers. In health care the pro-

gramme supported the introduction of a Community Health Fund, which is a

district/ward based insurance system. Members of the Fund contribute a certain amount

per year, for which they will be entitled to a minimum package of preventive and curative

out-patient services at dispensary and health centre level. The system was introduced in

1999 in all 22 wards of the district, but no data are yet available on participation in the

system or on its functioning. The programme also introduced an integrated approach

towards educational improvement, combining the rehabilitation of infrastructure,

teacher upgrading, the establishment of school committees and improvement of the

113



school inspection system. Finally, the programme started a pilot in one of the wards with

sub-contracting development activities to a ward-level Development Association.

The Association acts as a intermediary between the villages in the ward and the council,

including the financial management of funds made available by the Council (Songea

DRDP, Annual Progress Reports 1999; 2000; 2002). Not all these approaches have been

successful. The credit system is not functioning due to a very poor loan recovery rate and

attempts to rectify this have so far not been effective. Also, the pilot on sub-contracting

activities to a Development Association failed due to financial mismanagement.

Available information indicates that coverage of the various activities supported under

SODA/DRDP differs substantially. If judged by the number and percentage of villages,

coverage is sometimes high. For village and ward level planning, full coverage was

achieved, but for land use planning only 20% of villages have been covered. At the farm

household level, a relatively high proportion of farmers were involved in cashew tree

planting and, due to the spillover effect, also in improved irrigation techniques for paddy

growing. Substantially lower are the numbers involved in dairy, mechanisation through

animal draught power, and coffee growing. For income-generating activities an estimated

15,000-20,000 farm households have been reached, equal to about 20-25% of all house-

holds, but this includes the 11,000 farmers who received seeds for cashew nuts, the effects

of which are not yet known. Therefore, a more realistic figure is a coverage of 10-12%.

With regard to community services, DRDP ensured a 10% increase in water supply cover-

age, rehabilitated 40% of dispensaries and some 10% of available classrooms.

However, little is known of the effects of the programme in terms of increases in yield

levels and farm incomes, school enrolment, performance of teachers and quality of edu-

cation, and improvements in health status and nutrition of the population.


Progress and evaluation reports on Songea DRDP contain little information with regard to

its contribution towards poverty reduction. Only in the annual report for 2002 is reference

made briefly to this principal objective: “Despite efforts for more than 10 years to reduce

poverty in Songea Rural, it is widely assumed that poverty of farmers in general did not

become less. Although some of the reasons for this are clear, like long transport lines

from Songea to the main population centres in Tanzania and a poor marketing system,

other factors were not so clear to the district.’’ The internal evaluation report on SNV sup-

port to three districts also briefly touches on the issue. It concludes (i) that it remains dif-

ficult to judge whether the support contributed sufficiently towards improving the living



114

conditions of the rural population, and (ii) that a sustainable improvement of those living

conditions is not achieved through better social services but as a result of growth of the

local economy (SNV, 1997).

Available statistics indicate an increase in the number of schools and classrooms, but this

improvement of infrastructure was not accompanied by a higher gross enrolment rate,

chiefly due to an increase in population and a decline in the proportion of girls attending.

The sharply increased number of pupils selected for secondary schools (well-balanced

between boys and girls), however, shows an improvement of the quality of education.

Relevant figures for changes in the health situation could not be obtained. Moreover, it is

not possible to attribute improvements directly to the DRDP.

Perceptions about the interventions of DRDP in Songea differ among the various stake-

holders. The authorities view the question of poverty reduction in terms of improving

production, including the introduction of alternative crops, and acknowledge the serious-

ness of the marketing problem. Therefore, income-generating activities are perceived

favourably. The goals of poverty reduction are regarded by authorities and donors in

terms of targeting particular groups on the assumption that the market will take care of

the rest. Hence, the view that what is required is a partnership between government, the

private sector and NGOs. In reality, there is an extremely weak relationship and great

suspicion between the District Council, the private sector and NGOs. They not only

compete for the same sources of funding, but neither the private sector nor the NGOs

have sufficiently well-established organisations to allow for structural co-operation.

While support to the private sector resulted in the establishment of a Chamber of

Commerce, nothing similar happened with the NGOs. In addition, they suspect one

another with regard to technical capabilities, commitment and corruption.

The perceptions of government staff and elected leaders have been recorded also in the

self-assessment that preceded the mid-term review of 2000. Government staff mentioned

the positive contribution of DRDP to the development of the district. According to the

report, they stressed the departments’ commitment to the objectives of social and eco-

nomic improvement for the well-being of the population, in spite of inadequate funding,

inadequate and unmotivated staff, lack of reliable transport and lack of working facilities

(ETC, 2001a).

The District Council agrees with the support given to income-generating activities, as is

shown by by-laws that the council introduced on minimum areas of certain crops and the

115



introduction of alternative crops. At the level of the council and district officials, DRDP is

criticised for creating an invisible parallel structure rather than integrating its activities

and democratising local structures. According to a district official:

Most donors who come here have their own interests. They are willing to invest in things that they

are interested in. But if you make a proposal that you think is needed, you are told that the pro-

gramme is not ready to shoulder that. They came and said they were assisting the district council

in the process; they mostly dealt with those issues which they considered to be important from

their point of view… If they are not interested in a project, it dies a natural death…. When things

are not going well, their usual threat is they will withdraw their funding (Chachage et al,

2001)

In the view of the population, the projects aimed at improving social services and physical

infrastructure addressed the questions of poverty alleviation. They mentioned that these

services have greatly improved over the years, but that they did not necessarily solve the

problems of poverty. In their opinion, the serious problems are those in the sphere of pro-

duction and income.

There is no problem with the provision of water services… There are, however, serious economic

problems. People are poor. The (Tobacco)Board has made the people poor… The classifiers of the

Board want to be bribed Tshs. 2000,- per bale…. If one complains about this, he is told that you

do not want good economy (Chachage et al, 2001).

In the villagers’ opinion, DRDP’s support for income-generating activities does not

address the real problems. For example, credit facilities do not reduce poverty as the

amount provided is too small to have any impact and the costs are too high. Also, they

consider the dairy development project as not relevant for poor households because of the

high costs and risk involved, the long initial period in which it does not provide income

and the high labour input required. For the villagers it is not the projects that can help to

reduce poverty but the creation of conditions that could enable them to engage in pro-

duction and market their produce at fair prices.

There is also widespread criticism of the Village Level Planning effort, on the part of

villagers and council. It is seen as a bureaucratic exercise and a method of executing

projects rather than as a means to democratise village structures. The Village Assembly,

the main democratic institution is seen as being by-passed in decision-making, while the

committees created as a result of the introduction of projects essentially remained out-



116

side the village structures. Participation levels are also considered low, with some 25 to

30% of households attending meetings.


SODA/DRDP has undertaken a wide range of activities in four main fields: the improve-

ment of physical infrastructure, strengthening of the economic base in agriculture and

livestock, up-grading of social services for the provision of basic needs, and strengthen-

ing of the district administration. In the initial years most expenditure was on physical

infrastructure and social services, but in the second phase the emphasis shifted to

improving the performance of the district administration and strengthening the econo-

mic base, while maintaining the expenditure level for community services. Quite a

number of activities in the sphere of agricultural production were directed towards

diversification of the farming system and increase of productivity, and as such addressed

crucial rural development problems.

The programme introduced innovative approaches in agricultural credit through con-

tracts with village governments instead of individual farmers, in health care by establish-

ing a community health insurance system, and in education by introducing an integrated

improvement approach coupled with an effective school inspection system. Not all

attempts to introduce these innovations were successful, mainly due to weaknesses in

financial management and corruption.

Over time, the number of projects and project activities increased, resulting in a frag-

mented programme. Each of these may be relevant to a specific group or area, but it also

results in a relatively low number of beneficiaries per activity. This fragmentation is

caused by lack of analysis in the formulation stage of the programme to identify the cru-

cial bottlenecks in rural development in the district. The approach in village level plan-

ning, whereby communities were asked to list their development problems and priorities,

has contributed to this fragmentation. A thorough analysis would have revealed serious

deficiencies in agricultural services, marketing and input supply in particular. Moreover,

the apparent deficiency in genuine participation in village level planning may explain the

absence of marketing and input supply issues for main crops in a priority setting.

This led to the propagation of new crops such as coffee, cashew nuts and coconut instead

of tackling the crucial bottlenecks in dominant sources of agricultural income: maize and

tobacco. The zonal resource profiles were useful for appraising the feasibility of new

117



crops, but were apparently less adequate for identifying the main causes of low land and

labour productivity, the potential for technological innovations (except in paddy grow-

ing), and solutions to crucial problems in maize and tobacco production. Consequently,

the main issue in the perception of the population, i.e. deficiencies in the marketing of

maize and tobacco, has not been addressed by SODA/DRDP.

A crucial reason for not doing so may well be the limited scope for solving such problems

at the district level. Supervision of the marketing process to prevent bribery and corrup-

tion and achieve more transparency about prices and grading systems can be addressed

at the district level. However, general policy issues concerning effective liberalisation,

increased competition among marketing organisations and the reduction of political

interference, must be dealt with primarily at the national level. Embedding the district

development programmes in a sector-wide approach provides a link with the Ministry of

Regional Administration and Local Government, which does not have any authority over

agricultural marketing. The sector approach, therefore, is ineffective for addressing the

crucial causes of poverty at the macro level.

In spite of these problems and conflicts in agricultural marketing, production figures

indicate a fluctuating but gradually growing output of main crops and even increasing

gross returns. How does this relate to the people’s perception of declining farm and

household incomes and increasing poverty? Several factors may be responsible for the

difference: the limited reliability of production figures, the higher production costs for

farmers as inputs are no longer subsidised, and higher levels of taxation on crops. This

may have reduced farmers’ net incomes. Moreover, the perception study was carried out

in 2000/01, a year when a serious conflict in tobacco marketing occurred. For other

reasons, too, perceptions may not reflect the actual situation: the criticism on the dairy

project and the strong demand by people to be included in the project is a case in point.

In the absence of sufficient concrete and reliable information, the effectiveness of the

programme is hard to establish. It is not known whether road rehabilitation led to the

opening-up of new areas for agricultural production. It is not clear whether the propaga-

tion of new crops such as coffee, coconut and cashew resulted in higher net farm

incomes. Also, the profitability of innovations in animal husbandry has not been presen-

ted in progress reports. It is only mentioned that improved irrigation in rice cultivation

brought a six-fold increase in paddy output. In view of the spontaneous adoption of the



118

new technology by a substantial number of farmers, noticeable increases in farm income

may be expected

It may be assumed also that the support to social services resulted in better living condi-

tions in terms of better health and education facilities, and safer drinking water at shorter

distances for a larger proportion of the population. Apart from drinking water supply,

however, there is little concrete evidence on better quality of services.

In terms of capacity building and institutional development, the information is somewhat

contradictory. On the one hand, progress reports refer to large numbers of council and

departmental staff that have been trained and the good performance of the district in the

local government reform programme. On the other hand, perceptions of the population,

the District Development Plan and several evaluation reports refer to serious deficiencies

in leadership, in participation of the population in the planning process and, especially,

in financial management.

119





120

11 Assessment of drdp’scontribution to poverty alleviation

11.1 Socio-economic change in rural Tanzania

What were the essential processes of change in Mbulu and Songea that were relevant to

the present poverty situation? At the beginning of the twentieth century, the areas that

are now Mbulu and Songea Districts were characterised by self-sufficient agricultural sys-

tems with incidental sales of surpluses. In Mbulu livestock keeping was important next to

arable agriculture; in Songea arable agriculture was dominant. The commercialisation of

agriculture started in the 1940s. In Mbulu, Iraqw migrants embarked upon mechanised

wheat cultivation following the example of white settlers, in the 1950s followed by coffee

and pyrethrum growing. In Songea, farmers started to cultivate tobacco in the late 1940s.

However, agricultural commercialisation expanded in earnest after Independence: in

Mbulu in the early 1960s (wheat in particular) and in Songea in the 1970s (maize and to a

lesser extent tobacco). The principal factors responsible for that expansion were initia-

tives taken by entrepreneur-farmers and a stimulating government policy, the subsidised

provision of fertilisers, and compulsory minimum acreages for tobacco cultivation. This

process of change occurred in those parts of the districts with good soils and a favourable

climate.

The main effects of this agricultural commercialisation were an increase in farm income

and a growing social and geographic differentiation. Farmers used their higher incomes

for the improvement of their houses and for buying consumer goods. In general, and with

the exception of mechanised wheat farming, there was little investment intended to

increase productivity in farming. The growth of agricultural output was chiefly the result

of expansion of the cultivated area.

This process of output growth started to stagnate in the late 1970s and early 1980s, when

farmers were increasingly confronted with deficiencies in the state-controlled marketing

system and an over-valued currency. The economic crisis forced the Government of

Tanzania, among other measures, to devaluate its currency and to liberalise the marke-

121


Assessment of drdp’s contribution to poverty alleviation

ting system, including termination of the pan-territorial pricing system that favoured

marginal areas such as Mbulu and Songea. Agricultural production increases following

economic reforms were of short duration: in the early 1990s, subsidies on inputs were

gradually abolished and input supply deteriorated. The market did not compensate for

declining public services, as there was no genuine competition among providers of such

services. The effect was a decrease in land productivity and net farm incomes.

The decline of the rural economy was most serious in Mbulu District (in its present

boundaries), where a high rate of population growth in combination with marginal

climatic conditions led to land pressure, deteriorating soil fertility and periodic famine. In

Songea, where the natural resource position is more favourable (higher rainfall, a low

population density, a higher average farm size and only 10-20% of the land under cultiva-

tion, but with soils of rather low fertility), total output stagnated. Both areas suffered

from inadequate supporting services for the agricultural sector, input supply and marke-

ting in particular. This caused uncertainty among farmers about the prospects for long-

term agricultural growth and discouraged investments. With the exception of wheat

growing in Karatu and the recent innovations in paddy cultivation in Songea, there was

hardly any technological breakthrough resulting in higher land and labour productivity.

This history of economic change fits well into the overall picture for Tanzania. After the

country became independent in the early 1960s, the government initiated a rural develop-

ment policy which aimed at agricultural commercialisation with wide participation of the

population, intensive state intervention in, and control over, production, and heavy subsi-

dies on inputs. The latter was facilitated by fertiliser grants of donors. Several internal and

external factors caused the decline of the Tanzanian economy. Donors pressed for

economic reforms that were implemented from the mid-1980s onwards. These implied,

among other things, a reduction of state intervention, reform of the public sector, aboli-

tion of subsidies, and a return to a market-oriented economy with a prominent role for

private enterprise. Initially, economic reforms brought an increase in agricultural output,

that of food crops for domestic consumption in particular. However, further output

growth was hampered by low world market prices for export crops, inefficient marketing

systems and other supporting services, and stagnant productivity levels of land and

labour.

Social services in Mbulu and Songea improved after Independence, more or less in

conformity with the national pattern, but started to decline in the 1980s. Expansion in



122

services lagged behind population growth and their quality declined due to lack of

maintenance and shortages of staff and equipment. The decline was serious in Mbulu,

especially in terms of drinking water supply. In Songea, where the decline in social

services was less dramatic, there have also been more visible improvements since the

early 1990s. The question is: how did these socio-economic changes affect the present

poverty situation?

11.2 Poverty and perceptions of poverty

There are no accurate and reliable figures on household income and poverty in Mbulu

and Songea. For Mbulu the average per capita income was estimated at US$ 250 per

annum in the mid-1990s (i.e. including the present Karatu District), when an estimated

40-50% of the population earned less than required to satisfy basic needs. Moreover, the

area suffers from periodic famine. The average per capita income reported for Songea,

US$ 100-250 in the late 1990s, is much higher than the per capita value of main crops.

Improvements in real income after Independence, have stagnated since the late 1970s

and declined after the mid-1980s.

In the people’s perception there has been increasing poverty in Mbulu since the mid-

1980s, mainly manifested in low food security and deteriorating social relations. They

consider government’s neglect of agro-support services, weak local leadership and the

inadequate performance of the district administration as the basic causes of their present

situation. In Songea also, there is general agreement among the population about the

deterioration of their living conditions. In their opinion, poverty is manifest in low

incomes and the inability of households to afford social services, which are no longer

free, and to improve their houses. As the main causes of their present situation they see

the lack of access to agricultural inputs and the absence of reliable markets for their

produce, and weaknesses in leadership and organisation to address the people’s basic

development constraints.

National-level income data for Tanzania also point at persistent poverty and a decline in

living conditions since the mid-1970s. Present real incomes are below those of 1975. Low

world market prices for the main export crops and ineffective government policies have

been identified as the principal causes for the lack of economic growth and the stagna-

tion of incomes. As the production of food crops for domestic consumption (and exports

to neighbouring countries) prevails in Mbulu and Songea, deficiencies in government

policy seem primarily to be responsible for stagnant and declining levels of agricultural

123



output. Moreover, high levels of productivity would make farmers more competitive in the

market and therefore less vulnerable for price fluctuations. It is striking that, during

periods of higher prices in the past, farmers invested little in mechanisation and other

means with which to increase land and labour productivity.

The stagnation and decline in income in Tanzania is reflected in the people’s perception

of the basic causes of poverty. In two participatory poverty assessments during the late

1990s, the poor did not define the basic causes in terms of income. Instead, they empha-

sised socio-political factors as responsible for rural poverty and focused on their sources

of income and the services required to make effective use of those sources: access to and

control over land, social insecurity, and deficiencies in productive and social services.

These perceptions largely correspond with those in Mbulu and Songea: here also, defi-

ciencies in productive services (inputs and marketing) and deteriorating social security

were stressed. People in these districts also referred to weak leadership, low quality of

government administration, and deficiencies in the functioning of the legal system to the

detriment of the poor.

With the benefit of hindsight, it must be concluded that Tanzania’s policies for poverty

reduction have not been successful. After Independence, the Government of Tanzania put

high priority on poverty alleviation: poverty was labelled as one of the three “enemies of

development’’. Up to the 1980s these policies were characterised by heavy state interven-

tion in economic and social life and a strong focus on the expansion of freely-provided

social services, such as education, health and water supply. As state control over the

economy did not bring the expected benefits in economic growth and income increases,

social services could no longer be maintained and living conditions started to deteriorate.

The subsequent economic reform policies lacked any explicit poverty focus and the

presumed advantages of economic liberalisation did not materialise for large segments

of the population. Only since the mid-1990s, and encouraged by donor incentives for debt

reduction, has the Government of Tanzania formulated a comprehensive poverty

reduction strategy. This has been worked out in detail for rural areas and the agricultural

sector, and underlines the importance of increases in productivity and profitability in the

agricultural sector and the need for decentralised development planning. These are

crucial steps compared to earlier policies for rural development. Policy implementation

has hardly started, however, it confronts institutional weaknesses at lower levels of

administration, and is almost wholly dependent on donor funding. Its effectiveness for

poverty reduction therefore remains to be seen.



124

11.3 Local governance and poverty reduction

The assumption that decentralisation contributes to rural development is based on

considerations of better adjustment of development plans to the resource potential of the

district and greater influence of the local population in priority setting. This would

stimulate local participation in implementation of the plans and thus reduce costs for

government. How did decentralisation and local governance policies influence poverty

alleviation in Tanzania?

Tanzania’s socialist development policies of the late 1960s and the related concentration

of powers in the state brought an end to a gradual expansion of local governance which

had started in the 1950s. Characteristic for this period of socialist policies was the inte-

gration of the political and administrative systems. At all levels, the function of party

chairman and head of administration was in the hands of one person. And the party was

organised down to the households grouped into Ten-House Cells, each with its own

chairman. A rather limited form of decentralisation was introduced in 1982, including the

formal devolution of powers to elected bodies, although under strict supervision of the

ruling party. As it was widely acknowledged that institutional weaknesses at lower levels

of the administration might hamper effective decentralisation, capacity building was

considered an urgent requirement.

After introduction of the multi-party state in 1992, a more substantial reform programme

for Local Government started. Central Government committed itself to supply adequate

funding for the appropriate functioning of local administration, and created a basket

fund to facilitate donor support. In practice, the former socialist party continued to con-

trol the decentralisation process and it still dominates the district council in almost all

districts. In the sphere of local governance, Tanzania still appears to be a one-party state.

In Mbulu and Songea the people’s perception of local leadership and district government

reflects the institutional weaknesses. They consider the present leadership not to be inter-

ested in their main problems, and accuse them of a paternalistic attitude, misuse of

funds and corruption. Villagers claim that if any real improvement occurred in their lives,

it was in the second half of the 1970s. Despite the centralist system, they see that period

as one of good leadership.

These perceptions are confirmed by recent reports which point to councillors lacking

facilities to visit their electorates, the absence of women among elected members of the

council, and councillors who do not represent the people as envisaged. These reports also

125



refer to cases of malpractice and corruption and state that problems are not solved

because villagers fear their leaders. Finally, the reports mention that the effectiveness of

departmental staff is hampered by frequent transfers, unclear job descriptions, promo-

tions not based on performance, and serious deficiencies in financial management.

It must be concluded that, ten years after introduction of the multi-party system in

Tanzania, local government reforms have not yet resulted in the democratisation of local

structures in Mbulu and Songea, the generation of countervailing power in particular, nor

in structural improvements of district administration and the creation of an enabling

environment for the reduction of poverty.

11.4 Principal characteristics of district development programmes

The district development programmes in Mbulu and Songea are characterised by

long-term support of the Netherlands: since 1987 in Mbulu and since 1993 in Songea.

Assistance has been provided to a broad range of activities. Moreover, both programmes

have expanded support to increasing numbers of activities during the 10-15 years of their

existence. This has emanated in a complex and rather fragmented programme.

Both programmes had rather long periods of preparation, and Mbulu also a pilot period

with starter activities, but they lacked a baseline study which included a poverty analysis,

the identification of main obstacles for rural development, and the potential and required

strategy for improvement of living conditions and the alleviation of poverty. Implicitly, it

was assumed that the selected activities would contribute towards improving living con-

ditions in the districts; until recently, however, the correctness of that assumption has not

been tested by effective monitoring and evaluation.

During the preparation stage little or no attention was given to the policy environment in

Tanzania for the support of decentralised development activities. In retrospect, it may be

concluded that when the DRDPs started in the mid-1980s, Tanzania neither had a poverty

alleviation strategy nor a genuine decentralisation policy. Consequently, the policy

environment for such support was unfavourable. Over time, that policy environment has

improved, at least insofar as policy formulation is concerned, but concrete changes at the

local level are implemented only slowly.

Improvements in decentralisation policy have affected project organisation. Initially, the

programme was characterised by parallel structures, separate funding arrangements, and



126

the preparation of activities by a project office with a team of expatriate technical staff.

Recently, the programmes have been integrated wholly into the district administration,

and technical assistance has been reduced to a single advisor. Funds provided by the

Netherlands are earmarked for individual districts and provided on the basis of approved

district plans and arrangements about contributions from local revenue. This organisa-

tional change reflects a transfer of responsibilities to Tanzanian government staff and

councillors with the intention to contribute to more local ownership.

The two DRDPs had a low level of expenditure in per capita terms, i.e. 1-1.5 US$ per

annum. It is doubtful whether a serious reduction of poverty may be expected from such a

level of expenditure. Conversely, given the present institutional bottlenecks, higher levels

of aid could in all probability not have been absorbed. Moreover, the programmes did not

make an appropriate analysis of poverty and its underlying causes in the districts, sup-

ported a broad range of activities spread over the district instead of focusing on strategic

interventions, and changed priorities in the course of implementation. This reduced their

potential impact on poverty reduction.

The distribution of expenditure over types of activity reflects the changes in priorities and

strategy, although somewhat differently for the two districts. In Mbulu DRDP, rather high

levels of expenditure for physical infrastructure and agriculture in the initial period

decreased to the benefit of social services. Throughout the programme, various types of

training focused on capacity building at different levels of local government, but there

was no visible increase in expenditure for this cluster of activities. Also in Songea, expen-

diture on the physical infrastructure such as roads and bridges was dramatically reduced,

but expenditure on social services also fell. In contrast to Mbulu, Songea DRDP showed

an upwards trend in funding for income generating activities in agriculture and livestock,

and recently also for capacity building for good governance. It is difficult to assess the

relevance and meaning for the districts of these trends in expenditures, without taking

account of support given by other donors and of contributions by the Netherlands via

sector programmes for education and health. It is clear, however, that the changes were

the effect of a shift in priorities in Dutch development co-operation policy and of the

preferences of local politicians and district departments, rather than that they were based

on any analysis of needs and priorities on the part of the population of the two districts or

on a long-term strategy towards poverty alleviation.

127



Why did the Netherlands shift the emphasis in its support to DRDPs from strengthening

the economic base (increased incomes through higher agricultural production) to social

services and subsequently to capacity building of local government?

Several factors played a role. In the early 1990s, Dutch development co-operation placed

more emphasis on the social sectors and on ownership by the recipient country, and

gradually also on institutional development and good governance. The tendency towards

more ownership offered favourable opportunities to local politicians and district staff to

press for support to social services, which they considered tangible signs of progress.

Moreover, economic reforms encouraged bilateral aid to support those sectors which

remained in the public domain, i.e. health and education, a process facilitated by the

presence of sector specialists in the Netherlands Embassy. The priority for a sector-wide

approach in Dutch development co-operation strengthened the position of social sectors

and facilitated the incorporation of DRDPs into sector support for local governance. In

this way, a programme initiated for supporting the increase and sustainability of agricul-

tural production and productivity as a means to raise rural incomes, gradually turned into

a programme to stimulate decentralisation and strengthen district administration, on the

assumption that this would contribute to poverty alleviation.


The programmes supported four broad types of activity:

(i) construction and rehabilitation of physical infrastructure (roads, bridges, go-downs);

(ii) up-grading of social services by means of the rehabilitation of facilities and training

of staff to improve the functioning and operations of facilities;

(iii) strengthening of the economic base by the propagation of new crops and livestock

improvements, combined with environmental conservation and land use planning,

and some technological innovations in production (high yielding maize varieties and

improved irrigation in paddy growing); and

(iv) general capacity building and organisational strengthening, both in the public

sector, the district administration in particular, and the private sector (new type of

co-operatives, farmers’ organisations).

Direct output of the supported activities has been reported for both districts in terms of

passability of roads, number of schools and classrooms rehabilitated, percentage of

teachers trained, number of farmers provided with planting material, etcetera. The most

prominent improvements in Mbulu were the adoption of better land and water manage-

ment including the use of high yielding maize varieties, the planting of fruit and coffee



128

trees, and the introduction of improved dairy farming. In Songea there was a spectacular

improvement of irrigated paddy growing and drinking water supply through the rehabili-

tation of traditional wells. Songea DRDP also introduced interesting organisational

innovations in the social sectors (Community Health Fund, renewed School Inspection

system). Improvements cannot always be attributed directly or exclusively to DRDPs,

however; they sometimes supported on-going government programmes and other donors

were also active in the two districts (e.g. Denmark in water supply improvements in

Songea).

There is little or no information on the impact of improvements, such as increased agri-

cultural production due to road improvements, the propagation of new crops, and better

health status and quality of education because of support to the social sectors. Moreover,

there seems to have been little effect on the capacity building of local government/district

administration staff or on performance of the public sector in the districts. In Mbulu a

recent self-assessment showed a negative outcome of fifteen years of DRDP support: little

or no downward accountability of councillors, low participation in planning, and most

departments do not meet their targets in capacity building. Moreover, villagers reported

several examples of misuse of DRDP funds. In Songea recent reports also indicate serious

deficiencies in the functioning of government staff, little confidence of the people in the

leadership at all levels and, despite some improvements, continuous weaknesses in

financial management. Finally, the village planning exercise did not lead to genuine

participation of the local population, and had to be terminated due to its low efficiency.

It has been replaced by ward level planning, which is expected to raise efficiency but will

make participation even more challenging.

Despite the number of villages and households that have been included in the support of

DRDPs, the percentage of the population covered by the programmes is rather low, espe-

cially for activities directed towards strengthening the economic base and increasing

income. In Mbulu this was estimated at less than 5% of all farm households, in Songea

about 10-12%. Coverage in the social sectors was usually higher. The main reasons for the

low coverage were the broad range of activities supported, the institutional weaknesses in

local government and, more district-specific, the low potential of productive resources in

Mbulu and the necessarily time-consuming innovative approaches in Songea. The low

coverage, of course, reduced the impact of the programme: spontaneous adoption of the

improvements by the population was the exception rather than the rule. Irrigation

improvement in Songea is an interesting case in point: a level of technology which could

129



be adopted by farmers without much external support, a substantial increase in land

productivity and a high return to family labour.

It is striking that several priority areas perceived by the population have been given little if

any attention in the programmes. This applies especially to deficiencies in input supply

and marketing for main crops, and the increase of land and labour productivity through

technological innovations. Also, they include the lack of social security due to the

absence of employment opportunities for youngsters in Mbulu and weaknesses in the

functioning of courts of law. There are two principal reasons for not addressing these

perceived priorities: the absence of any analysis to identify the main obstacles for rural

development in the preparation phase of the programmes, and the limited possibilities to

solve such problems at the district level. Several main obstacles to district development

fall outside the authority of district administrations and can only effectively be addressed

and solved at national level. The sector-wide approach and the incorporation of DRDPs in

sector support to local governance does not create the appropriate national framework for

problems of partial marketing liberalisation and ineffective marketing structures.


Stakeholders involved in the programme have different opinions about the impact of the

programmes with regard to improvement of living conditions and the reduction of pover-

ty. Government staff involved in the programmes are inclined to assess positively the

fruits of intervention; they attribute shortcomings to annual delays in the release of donor

funds, and to the conservative views and attitudes of farmers. Village leaders and council-

lors share the optimism of government staff about the role of development interventions.

Their main concerns are consistency and equity in the distribution of funds and the

tangible results of the support to enhance their credibility among their electorate.

Therefore, they give priority to the construction of physical infrastructure, and other

visible structures such as classrooms, water supply systems and health units; they

complain about money spent on training courses and seminars and about DRDP’s strict

requirements for planning and programming.

Most ordinary villagers do not distinguish between donor and government interventions,

also because all funding is channelled through government departments. When they do,

they see government personnel as hampering the flow of development resources to the

population. The rural population appreciated projects that improve social services, but

claimed that these improvements did not solve their poverty problem. In their opinion,



130

the serious problems are those in the sphere of production and income. DRDP support

does not address their real problems, i.e. the creation of conditions that would enable

them to engage in production and to market their produce at fair prices. There is also

wide-spread criticism of village level planning, which they see as a bureaucratic exercise

with all decisions taken by government officials. They consider it a method of project

preparation and implementation rather than a means to democratise village structures.

The critical attitude of the population was undoubtedly influenced by specific conditions

at the time of the interviews: famine in parts of Mbulu and a crisis in tobacco marketing

in Songea. Moreover, criticisms ventured in the perception study did not always corre-

spond with the interest shown by farmers in agricultural programmes, such as dairy and

paddy improvements. Nevertheless, opinions about DRDP support and the role of govern-

ment in the planning and implementation of development policy point to fundamental

differences between the people and their leaders about the types of intervention required.

The overview of activities and output shows that the two programmes contributed to

improvements of the road system, the propagation of new crops and new systems of ani-

mal husbandry, the rehabilitation of facilities for social services and the training of their

personnel, and the introduction of new forms of social organisation to serve the interests

of the local population. Achievements in Songea seem to have been better than in Mbulu,

partly due to the better resource potential, but also because of greater attention for

strengthening the economic base. Diversification of the cropping pattern as pursued in

Songea, and exploring the possibilities for technological innovation and raising the

profitability of farming, are in line with the perceptions of the population: creating an

enabling environment for agricultural production increases. The weakness of both

programmes was that they did not address the deficiencies in support services for main

crops as a pre-condition for any further improvement in the agricultural sector. To what

extent did these activities result in a structural alleviation of poverty ?

Progress and evaluation reports of the two district programmes contain little concrete

information about their contribution to poverty alleviation. For Mbulu, the low coverage

of households for income-generating activities and the absence of targeting on the poor

in DRDP interventions indicate little impact on poverty reduction. For Songea, the most

recent progress reports states that the poverty of farmers has not been reduced in spite of

ten years of effort to that end. This corresponds with the perception of all parties in the

districts that poverty has been increasing since the 1980s. Moreover, the assumption of a

poverty reduction effect from the strengthening of local government has yet to be proven.

131



In the perception of the population the increase in poverty is partly the effect of a rigorous

taxation scheme the Council is implementing in relation to DRDP policy of enhancing the

council’s financial capacity. Paradoxically, the attempts to increase ownership and sus-

tainability of Dutch support to DRDP have had an unfavourable effect on poverty reduc-

tion. Similar observations have been made for fiscal decentralisation in Uganda. In the

early 1990s, however, the World Bank emphasised the importance of lower taxation of

agriculture as a means to encourage rural development and successfully reduce poverty.

It must be concluded, therefore, that the two DRDPs have not yet achieved their main

long term objective, structural poverty alleviation, nor their short-term objective of

strengthening local governance. If DRDPs continue in the present way, their contribution

to poverty reduction will be negligible also in future.

The question is to what extent these conclusions are relevant for the overall Dutch sup-

port to district rural development in Tanzania. Available information does not permit a

straightforward answer to that question. Although the programmes have been evaluated

several times, these evaluations either covered only some of the fourteen districts includ-

ed under Dutch support or, if all districts were covered, as in the mid-term review in 2000,

did not contain any information about the impact on poverty reduction. Those evalua-

tions that did refer to poverty reduction mention the absence of a poverty analysis, the

little attention for productivity increases in agriculture, and the marginal contribution to

direct poverty alleviation. Furthermore, it has been observed that the shift to capacity

building in local governance may conflict with the now secondary objective of creating a

conducive environment for increasing rural incomes.

The two districts included in the present evaluation are to some degree representative of

several others, i.e. those with adequate productive resources, providing that attention is

given to maintaining soil fertility (Songea), and those with low potential and relatively

high population pressure (Mbulu). The DRDPs in these districts have about the same level

of expenditure, operate under similar conditions with regard to the macro-economic and

policy environment, face the same institutional weaknesses in local governance and are

also confronted with production conditions, whereby the crucial functions of a withdraw-

ing state have not been replaced by a well-functioning market. Taking account of

available evaluation findings, it may therefore be expected that the conclusions of the

present evaluation have wider applicability for Dutch support to district rural develop-

ment.



132

Finally, how are the conclusions of the present evaluation related to other evaluations of

support to decentralisation, local governance and poverty alleviation? A recent study on

decentralisation, commissioned by the OECD/DAC network on aid evaluation pointed

towards the weak poverty focus of many donor programmes, the little evidence that donor

interventions contributed significantly towards poverty reduction, and the lack of com-

mitment on the part of governments of recipient countries. The findings of the present

evaluation in Mbulu and Songea correspond largely with those of other evaluations.

The findings of the Bosuyt and Gould study in three countries in Sub-Sahara Africa are

particularly relevant as they analysed the effects of donor programmes in a similar socio-

political environment as that of Tanzania. Dutch support to district rural development

programmes corresponded with some of their recommendations, such as incorporation

of the programme in statutory structures in Tanzania and support to central agencies in

charge of decentralisation. Others were clearly not addressed, however. The main

deficiencies were: (i) support for decentralisation was linked insufficiently to poverty

reduction, (ii) there was little effort for political empowerment and strengthening of civil

society, in combination with attempts to improve public administration, or these efforts

were ineffective, and (iii) inadequate attention was given to resource mobilisation as an

instrument for income growth.

133





134

Annex 1 The Policy and Operations

Evaluation Department

The Policy and Operations Evaluation Department, in Dutch the Inspectie Ontwikkelings-

samenwerking en Beleidsevaluatie (IOB), is responsible for conducting evaluations of

Netherlands foreign policy.

IOB is part of the Ministry of Foreign Affairs. It is an independent unit, which reports

directly to the Minister of Foreign Affairs or the Minister for Development Co-operation.

The Minister concerned submits IOB reports accompanied by a letter with his/her policy

reactions, to Parliament, where they are discussed by the Permanent Committee on

Foreign Affairs with respect to follow up actions.

IOB was established in 1977 with a mandate that was restricted to the evaluation of aid

programmes. Following the reassessment of Netherlands foreign policy in 1996, IOB’s

mandate was broadened to include other fields of foreign policy.

From 1977 to the mid 1980s, IOB’s emphasis was on individual project evaluations, the

status of which was then confidential. Since the mid 1980s, emphasis has shifted to com-

prehensive thematic studies, focusing on policies and modalities of implementation and

covering sectors, themes or programmes. External independent experts participate in the

various phases of the research, under the responsibility of IOB. Where relevant, institu-

tions or experts in recipient countries are invited to participate in the fieldwork. In some

cases reference groups consisting of independent experts and Ministry staff are appoint-

ed for the evaluations, to advise on the methodology, approach or subjects under review.

The final reports, based on various field and desk studies, are published under the

responsibility of IOB.

In addition to its own evaluations, IOB also participates in multi-donor evaluations.

135


Annexes


Annexes

136

Annex 2 Terms of Reference location

specific poverty study

1. Reasons

During the departmental theme meeting of 20 April 1998, discussions were held with the

Minister for Development Co-operation regarding the internal memorandum on poverty

and development. Fundamental questions were asked about the effectiveness of activities

intended to reduce poverty. It was said that the Ministry’s capacity to learn should be

given greater focus and that a general approach should be adopted to ensure effective

co-ordination with national authorities, other donors and non-governmental organisa-

tions on the issue of poverty reduction. Parties were dissatisfied about the relationship

between policy and implementation, feeling that there was a tendency to lose sight of the

objective.

Evaluations have devoted little specific attention towards establishing whether activities

funded by donor money have had any impact on poverty, neither has the matter been a

focal issue in evaluations conducted by the Ministry’s own Policy and Operations

Evaluation Department (IOB).

Country studies conducted by IOB in the recent past have included perception studies to

examine how Dutch projects have been experienced by target population groups.

Few conclusions were reached, however, about the relevance of the activities to poverty

reduction and it was only partly established what target population groups were truly

interested in.

This evaluation will take as its point of departure not only the problem of poverty but also

its causes and trends, and poor people’s own perceptions. The aim is to assess the

relevance, effectiveness and efficiency of interventions. In the long run, the evaluation

services of recipient countries and other donors could be asked to participate in follow-up

studies.

2. Introduction

Poverty is a relative term, and degrees of poverty are measured against indicators that

have been developed by governments of developing countries and by donors. Although

these indicators are used to justify interventions and the funding of activities, they do not

137


Annexes

necessarily reflect the views of the target groups towards whom support is to be oriented,

nor are they sufficiently based on analysis of poverty problems.

Poverty reduction has been the main objective of development co-operation efforts for the

past 50 years. Some interesting results have been achieved in the economic and social

sectors of developing countries. In the last three decades life expectancy has increased by

more than 20 years, child mortality has been halved, and twice as many people now

receive primary education. Average income, health status and literacy rates have

improved faster in developing countries in the past century than in the OECD countries

(DAC Guidelines on Poverty Reduction, Vol. I, June 2000 Draft).

The World Bank’s 1990 prognosis of a 3.9% annual reduction in poverty has not been

fulfilled. Instead, it has averaged only 1.5% a year in the last decade. In the years between

1990 and 1998 the proportion of the population in developing and transition countries

living on less than US$ 1 (in 1990, and US$ 1.08 in 1993) fell from 29% to 24%. If China is

disregarded, however, the number of poor people in these countries increased in absolute

terms by 70 million. Global income inequality rose accordingly. In all low-income coun-

tries, with the exception of China and India, income per capita dropped by 1.4% a year

between 1985 and 1995 and by 0.7% a year in medium-income countries (OED World

Bank 1999). National figures sometimes disguise internal differences. In a number of

countries, (e.g. Burkina Faso and Zambia) inequalities between certain population

groups – particularly women, rural populations and minorities – and the rest have

increased, despite national economic growth. Poverty is not static. Over a period, more

people’s incomes fluctuate above and below the poverty line than stay permanently below

it (Baulch Hoddinott 1999).

Apart from its effects on income, there are other reasons to conclude that the results of

donor support are inadequate. There has been a major shift in the definition of poverty

itself over the years. At first, comparisons were mainly based on per capita income. By the

early 1980s, meeting basic needs was already an accepted notion (P. Streeten et al. 1981).

UNDP developed a series of indicators by which to define poverty in the Human

Development Reports (UNDP 1996), the ILO adopted the social exclusion approach, and

Chambers involved the population itself in determining the indicators (R. Chambers

1995). To Amartya Sen (1999) development was linked to freedom of the individual.

Development thus implied the removal of all constraints on that freedom. This notion

dovetails with that developed by IDS, in which a pentagon is used to represent the


Annexes

138

development of the various types of capacity (D. Carney 1998), and with DFID’s livelihood

strategy, which allocates a central role to resources and services.

Expanding the notion of poverty has led to greater understanding of what it entails, so

that more and more objectives, criteria and norms are gaining international acceptance.1

The donor community is now seeking ways of measuring whether its objectives are being

achieved. The close interaction between activities and priorities on the one hand, and

developments at local level on the other, is often underestimated. As a result, evaluations

of poverty are often complicated.

The way in which people perceive their own situation is an important factor in any inter-

vention. Some of the changes that take place in families’ immediate environment are

autonomous in nature. Others are produced by all sorts of interventions, some of which

may be supported by donors. Individual family members respond in their own way to

these changes, which are location-specific and dependent on how the population expects

their situation to improve. These responses then influence the effectiveness of changes

envisaged in the situation of authorities and donors and the efficiency of efforts made.

This evaluation takes the problem of poverty and the people’s own perception of it as the

twofold basis of its assessments. This means that the study has to be conducted from the

reverse angle. It will thus encroach on virgin territory, adding a second complication.

3. Aims and set-up of the location-specific poverty study

IOB intends to carry out an evaluation of changes taking place in a specific area, and to

do so from the viewpoint of the population. The aim is to examine what these changes

mean to people, and to gauge the response to them of individuals and civil society orga-

nisations. This applies not only to autonomous changes, but also to those induced by the

authorities, local organisations and donors. The evaluation will also examine the

relevance of funded activities to the target group, to the policy of the country concerned,

and to donors. It will also assess the effectiveness and efficiency of activities that received

Dutch funding.

139


Annexes

1 Seven International Development Goals (IDGs) have been formulated for the year 2015: (1) to reduce the per-centage of the world’s population living in extreme poverty by half; (2) universal primary education; (3) to eliminate gender disparity in primary and secondary education; (4) to reduce infant and child mortality by two-thirds; (5) to reduce maternal mortality by three-quarters; (6) universal access to reproductive health services;(7) to implement national strategies for sustainable development. An 8th item was added during the 1995World Food Summit: to reduce the number of undernourished people by half (DAC Guidelines on PovertyReduction, Vol. I, June 2000 Draft).

These terms of reference set out the locations, the set-up of the various studies and the

methodology.

3.1 Locations

A number of options were considered in the preparatory study. Locations have been

selected for their relevance to current policy, and to enable comparisons to be made

within a similar socio-economic context. An effort has also been made to minimise the

risk that coincidental circumstances at the location will influence or determine the

indings of the study.

Eligible for selection were:

• Countries with which the Netherlands implements a regular development

programme (18+ countries);

• Countries in which poverty reduction plays a dominant role;

• Locations that form an administrative entity, which have not been hit by recent

disasters and whose population is not nomadic;

• Countries where two locations are eligible for selection, having received Dutch

support over a long period;

• Locations where the Netherlands has funded several activities.

Three African countries were eligible for selection on the basis of these criteria: Mali,

Burkina Faso and Tanzania. A number of evaluations have recently been carried out in

Mali, however leaving Burkina Faso and Tanzania as possibilities. The Tanzanian govern-

ment exerts considerable influence in rural areas, and decentralisation has strengthened

its influence at the local level. Tanzania was therefore selected. The Operations Evaluation

Unit (IOV) carried out a country study here in 1994, and much of the information

collected then will be useful now.

The Netherlands supports 14 districts in Tanzania. A number of these are populated by

nomadic peoples, while others provide shelter for refugees from neighbouring countries.

Mbulu district in the north and Songea district in the south were selected. A major reason

for selecting Mbulu was the extensive documentation that has been published on the

area. In addition, it has received Dutch bilateral aid for some considerable period. This

also applies to Songea, which was selected partly because the Dutch aid effort is in the

hands of the Netherlands Development Organisation (SNV). Information may therefore

be forthcoming from SNV’s other planning systems.


Annexes

140

In Asia, Sri Lanka is an attractive choice because two well-documented programmes are

being implemented there in similar socio-economic environments. Local research insti-

tutes in Sri Lanka have done extensive studies into the results of integrated rural develop-

ment programmes. The next location could be sought in Latin America. Experience with

the first two location-specific studies will determine whether more should be carried out.

The Tanzanian evaluation does not particularly seek to be representative. It is the first

study of its kind of the position occupied by induced changes in a society and uses

methods not previously applied by IOB to achieve this objective. A few constant factors

may possibly be identified once this study has been conducted at two locations. Future

studies could devote specific attention to the representative nature of the findings.

3.2 Study structure

Documentation will first be collected on the locations, including general information on

the socio-economic background and more specific information on regional development.

These data will then be analysed and the main trends and poverty indicators identified.

The main determinants that influence the processes decisive for development of the

districts will be identified.

Video interviews will be used to map out the main features of poverty from the point of

view of people living in the districts: how do they experience their own situation, what are

their main needs and what changes do they want to be brought about? Do people feel

that activities are relevant and, if so, how do they assess them? The video evaluation

method will be used in the study at the first location. In the light of the experience thus

gained, a decision will be made regarding use of this medium at the second location.

At the same time, a source study will be carried out into Dutch support over the entire

period. The principal elements will be planning processes, the reasons for setting priori-

ties, and related funding. The funding provided by the national government and by other

donors will also be analysed. This study should give rise to a list of the main objectives

envisaged by interventions in the district, and show how existing evaluation reports have

rated the success achieved.

During the field studies, the target population group will be asked to talk about changes

that have taken place in the past decades, and to give their opinion of them. The results

141


Annexes

will be compared with those of the source study and the regional analysis. The main aim

is to be able to draw conclusions about the relevance of interventions from the analysis of

poverty problems and from the viewpoint of the people involved. Where they are

considered relevant, further studies will be carried out to examine the effectiveness of the

changes. Where possible, efficiency will also be assessed.

3.3. Methodology

The regional study method will be applied to the districts on the basis of existing

documentation. The system to be used in conducting this study will depend on the

physical and socio-economic development trends identified in the districts

Video will be used on an experimental basis in a local study of the population of the dis-

trict, who will be asked to help produce the film. They will be given questions to be asked

in interviews which they will also record. The recordings will then be shown to other

people living in the same area for their comments and to verify the opinions expressed in

the interviews. From all this recorded material, a short video film will be produced

containing the results that are most relevant to the objectives of the study.

The source study will comprise various components. Material filed at the ministry will be

used to examine how Dutch support for the district has proceeded, and on what basis

funding has been allocated. Some of the documentation collected since 1996 has been

filed at the embassies, and will have to be analysed there. In addition, funding has gone

to a great many interventions for which the national government was not responsible.

Documentation on other donors will be collected and analysed in the country itself. The

information will be collated, and subjects that have an important bearing on induced

changes will be sifted-out.

The methods to be applied in the field studies will largely depend on the results of the

source study, needs identification and regional studies. The more information that is

acquired on the relevance of induced changes, the greater will be the chance of formula-

ting specific core questions for the field studies. These core questions will be defined, and

the methodology for the field studies worked out, in the course of the first studies, and

will depend on the progress made with them. The method that springs directly to mind is

that of participatory rapid appraisal in its various guises, but household studies and

incident observation and analysis are also worth considering. A seminar on methodology

will be held before the field studies are carried out.


Annexes

142

4. Stepwise evaluation

The three studies referred to under 3.2 (regional, needs identification and source study),

which will be carried out at the preliminary stage, should lead to the formulation of core

questions for the field studies and the evaluation as a whole. They will be identified by

comparing the population’s perception of their position and their priorities for poverty

reduction to the priorities set by government and donors for their interventions and to

analyses of regional development in the districts.

The first questions to be answered have to do with relevance. What were the main needs

and issues identified by the people in order to improve their own situation? What regional

development processes are essential to poverty reduction? What role do induced changes

play?

A second set of questions has to do with the effectiveness of induced changes in strength-

ening or possibly impeding poverty reduction processes. How can that be assessed?

Can any identified effects be explained partly by interventions’ differing relevance to the

various parties involved?

A third question has to do with the people’s perception of resource deployment and its

efficiency. Do the activities dovetail with regional development trends? Are the resources

allocated proportionate to the results expected?

For this part of the evaluation, field study methods will be developed to ensure an

efficient response to the questions. These methods will depend partly on the core

questions for this part of the study.

5. Evaluation, products and planning

5.1 The studies

A consultant will be commissioned to do the first study, i.e. the regional analysis. The

consultant will be responsible for collecting documentation on the macro- level socio-

economic and political development of the country on the district, and on the activities of

other donors and NGOs, and will produce a regional analysis on their basis. Finally, the

consultant will produce poverty indicators based on the trends and processes unfolding

in the districts.

143


Annexes

A specialist experienced in the use of video in carrying out local perception studies will be

contracted to do the second study, identifying the needs of the target group. It will be the

specialist’s job to establish which issues and priorities are relevant to poverty reduction

from the viewpoint of the people. Selected passages from the video recordings in which

local people pinpoint the issues they feel require attention will be used to make a video

film lasting no longer than one hour. The specialist will also produce a short written

report describing the process and results. The contract will cover the study in the first

location. If the results are satisfactory, it will be extended to cover the second.

The third component comprises a source study to analyse Dutch funding in the districts.

A researcher has already been commissioned for this work, which entails collecting

relevant information from the ministry’s records. These are proving to be incomplete, and

decisions about funding appear to have been taken without any kind of written justifica-

tion. Since the foreign policy review in 1996, a large number of records have been filed at

the embassies and the ministry knows little about them. The consultant responsible for

the regional studies will analyse the material now being collected by the researcher. Other

documents that need to be retrieved are those in the hands of the national government,

of other donors, and non-governmental organisations in the districts.

The results of these studies will form the basis of the field studies. It is vital that these

results are supported by the various stakeholders involved. A workshop will therefore be

organised in Tanzania to present and discuss the results obtained to date. The selection

of stakeholders and the methods to be used in the field studies will also be discussed.

It is difficult to give already now precise definition of the main features of the fourth

component, i.e. the field studies. The core questions to be asked here will depend on the

results of the previous studies. The main point at issue is the relevance and effectiveness

of induced changes. The target population is diverse, comprising men and women, young

and old, city and country dwellers, and people of varying ethnic origins, so that any field

study will involve considerable time and money. Both, however, are crucial in ensuring

the quality of the study.

Note: As explained in chapter 2 on the methodology applied in this evaluation, the objectives of the

Terms of Reference could only partially be executed due to shortcomings with respect to data-gathering

and the registration of perceptions on poverty of the local

population.


Annexes

144

Annex 3 Tables

145


Annexes

Table 3.1 Expenditure on DRDP 1987-2002 (000¤)

Year Bilateral SNV Total LGRP Alat Grand total

1987-92 1.320 – 1.320 1.320

1993 2.903 204 3.107 3.107

1994 3.536 269 3.805 3.805

1995 4.343 1.411 5.754 5.757

1996 6.214 1.570 7.784 7.784

1997 8.560 1.855 10.415 371 10.786

1998 8.747 2.020 10.767 727 11.494

1999 7.444 2.642 10.086 635 10.721

2000 7.110 1.236 8.346 1773 10.119

2001 5.290 1.184 6.474 392 122 6.988

2002 6.756 2.125 8.881 346 85 9.312

Total 62.223 14.516 76.739 4244 207 82.190

Total NLG 182.500.000


Annexes

146

Table 3.2 Major Activities Mbulu DRDP

Source: Mbulu DRDP Plan 2001

Income Generation

Crop promotion: training, demo plots, seedlings,

sunflower, pyrethrum, irrigation, horticulture, oxen,

storage, extension, research

Livestock promotion: indigenous, dairy

Natural resources protection: soil fertility,

agroforestry, seedlings, tree planting, training, wild

life conservation, village boundaries, urban master

plan

Group formation: savings/credit societies, village

management

Feeder road rehabilitation

Health

Improved health services: staff training,

renovation, radio communication

Disease reduction: health education, cold chain,

immunisation, water wells and tanks, user

groups, training

147


Annexes

Education

Improved education standards: teacher training,

monitoring framework, methodology, gender,

manuals, accounting, planning, inspection,

upgrade teachers, rehabilitate class rooms

Capacity Building

Training of ward executive officers, planning, computer

training, accounting

Community Participation: participatory planning work-

shops, extension workers and village development

facilitation teams

Land use planning: gender balance

Maintenance: fund, training,

District maintenance policy: plan, training, monitoring


Annexes

148

Table 3.3 Major Output MDRDP during 1998-2000

Source: Mbulu DRDP Plan 2001

Income Generation

Crop Promotion:

250,000 seedlings distributed, 40 plots sunflower, 20 plots

groundnuts, 500 kg pyrethrum sold to farmers, 441

members irrigation committees trained, 7 irrigation

structures built, 160 farmers trained in horticulture, 50,000

fruit seedlings surviving, 1800 pairs of oxen trained, 120

pairs of donkeys trained, 80 oxcarts constructed, 60 farmer

groups established, 120 cutworm trials, 11 storage structures

built, 578 farmers trained, 15,000 farmers trained in new

technology, 179 adoption plots established, 10 demo plots,

20 stockist operating.

Livestock Promotion: 171 heifers procured, 37 bulls, 220

acres pasture, 171 sprayers, 210 farmers trained, 72 pig

farmers trained, 3 dip committees formed, 9 paravets

trained.

Natural Resources: on soil management 1550 farmers

trained, 270 acres conserved, 40 km planted grass, 150 oxen

users trained, 16 km gully controlled, 13 villages involved;

150,000 seedlings raised, 55 demo farms, 1m tree seedlings

planted, joint forest management in 10 villages, 141 water

sources protected, poaching reduced in 9 villages, 124

village leaders trained in boundary survey.

Group Formation: 10 SACCOs operating, capital increase

from Tsh 8m to Tsh 22m, members from 400 to 2500, 6

bookkeepers trained,

Feeder Road rehabilitation: 14 villages more accessible,

13 new culverts, 39 people trained in maintenance

Health

Health Services: 700 staff trained (mainly

traditional birth attendants), new

prevention approach in 2 wards,

7 dispensaries renovated, 10 radios.

Disease Incidence: improved sanitation

skills in 9 wards, 3 refrigerators, mobile

clinics, 22 water wells, 7 water tanks,

30 water user groups trained.

149


Annexes

Education

Education Standards: 556 teachers trained,

pupils pass marks improved, 33 health education

trainers trained, 32 teachers trained in

environmental issues, methodology manual,

31 teachers trained in gender, 171 headmasters

trained in accounting, 52 schools inspected,

78 teachers upgrading, 173 improved class

rooms, 111 teacher houses,

Capacity Building

First steps bottom-up planning in 71 villages, 16 ward

executive officers trained, 24 staff computer trained,

new national accounting system introduced, 13 village

meetings on community participation, 13 extension

staff trained, 2 meetings on gender balance in land use

planning, 28 extension workers trained in

maintenance, 15 school committees formed, 9 water

user groups, 5 maintenance funds, 20 villagers trained

in maintenance, 5 maintenance accounts.


Annexes

150

Table 3.4. Activities and output SODA/DRDP,1994 - 2002

DEPARTMENT

COMMUNITY

DEVELOPM.

WORKS

LANDS

DESCRIPTION

Empowerment of

villages

Construction of

bridges

Construction of

roads

Participatory Land

Use Planning

(PLUM) in villages

1994

39 villages trained

on village level

planning (VLP)

1995

Same 39 villages

facilitated in VLP

and preparation of

VAP

5 bridges

constructed

60 km spot

improvement

3 villages (LUP)

1996

facilitated in VLP

and preparation of

VAP in same 39

villages

12 km Lusewa –

Ligera road

rehabilitated

1997

7 additional

villages trained in

VLP and facilitated

to prepare VAP

16 bridges

constructed

12 km spot

improvement

11 villages

(LUP)

OUTPUT ACHIEVED

151


Annexes

1998

17 additional

villages trained in

VLP and facilitated

to prepare VAP

4 bridges

constructed

12 km spot

improvement

- PLUM team

formed

- Materials/Equip-

ment purchased

1999

53 additional

villages trained in

VLP and facilitated

to prepare VAP

8 bridges

constructed

12 km spot

improvement

PLUM facilitated

in 4 villages

2000

PRA in all 116

villages, covering:

- Evaluation of VLP

- Ward

Development

Plans (WDP)

designed

6 bridges

constructed

12 km spot

improvement

PLUM conducted

in 4 villages

2001

Ward Plans

developed in all 26

wards

2 bridges

constructed

21 maintenance

group formed

PLUM conducted

in 5 villages

- SDC fixed assets

valued

2002

128 Village Govt.

members trained

on District

Planning Cycle.

3 bridges / culvert

constructed

28km improved

and

- 10 maintenance

group formed

3 villages

conducted PLUM

OUTCOME

- by 1999, all

villages in the

district covered by

VLP and VAP

- by 2001 all wards

covered by WDP

- Village Govt. and

Ward Dev. Com-

mittees trained

- participatory

planning process

introduced and

established

- by 2002 first

integrated district

development plan

Three quarters of

SDC roads are pass-

able throughout the

year as opposed to

25% in 1994

Maintenance groups

contracted by SDC

Reduced numbers of

boundary conflicts


Annexes

152


DEPARTMENT DESCRIPTION 1994 1995 1996 1997

AGRICULTURE

& LIVESTOCK

Cashewnut

Propagation

Coconut

Propagation

Cockerel Project

Piggery

Coffee

Propagation

Oxanization

5 villages provided

with cockerel

Additional 5

villages provided

with cockerel

5,612 seedlings

distributed to

farmers

12,000 seedlings

distributed to

schools

Additional 10

villages provided

with cockerel

110 pigs

distributed to

farmers in 4

Villages

33,360 seedlings

to selected

farmers

OUTPUT ACHIEVED

153


Annexes

1998 1999 2000 2001 2002 OUTCOME

32 acres planted

6,000 seedlings

purchased and

distributed to

farmers

Additional 2

villages provided

with cockerel

27 boars

distributed to

farmers in 3

villages

60,000 seedlings

distributed to

selected farmers

20 farmers trained

in 3 villages

Additional 400

acres planted

20,000 seedlings

distributed to

farmers

12,000 seedlings

purchased and

distributed to

schools

Additional 9

villages provided

with

500 cockerel

22 boars

distributed to

farmers in 3

villages

20,000 seedlings

distributed to

selected farmers

20 farmers trained

in 7 villages

10 villages

received cashew

seeds,210 hectares

planted

12,000 seedlings

purchased and

distributed to

farmers

Additional 16

villages provided

with cockerels

100 piglets

distributed in 4

villages

20,00 seedlings

distributed to

selected farmers

20 farmers trained

in 3 villages

Additional 10

villages with 210

hectares planted

12,000 seedlings

purchased and

distributed to

farmers

Additional 3

villages provided

with cockerel

40 piglets

distributed in 1

village

80,000 seedlings

distributed to

selected farmers

604 farmers

trained in 4

villages

Cashew study

started and

presented to the

Council

80,000 seedlings

distributed to

selected farmers

1104 farmers

trained in 3

villages

- total of 595

hectares planted

- recommendations

of study in process

of implementation

Schools sell

coconuts for school

funds

Improved chicken

stock in 50 villages

Increased income

for the farmers

299 farmers

received pigs/boars,

directly benefiting

about 1,350 people

Increased income to

the farmers

Increase acreage

being cultivated,

generating more

income


Annexes

154



TRADE

WATER

Introduction of

Irrigation

Techniques for

paddy production

Heifer Intrust

Scheme

Establishment of

Village and Mobile

Markets

Water Hand

Pumps

Construction of

Traditional Water

Wells

—

50 heifers and 5

bulls distributed to

50 families in 3

villages

— 10 villages

provided with

handpumps

50 heifers and 5


50 families in 5

villages

24 pumps

purchased for 4

villages

105 Traditional

Water Wells

improved 6

villages

OUTPUT ACHIEVED

155


Annexes

1998 1999 2000 2001 2002 OUTCOME

50 heifers and 5


50 families in 5

villages

21 handpumps in

5 villages

55 Traditional

wells improved in

3 villages

20 farmers from 3

villages identified

and trained in

Irrigation Tech.

Introduced

110 Heifers and 11

Bulls distributed

to 110 families in

11 villages

2 markets

constructed at

Peramiho A &B

32 handpumps to

8 villages

109 traditional

wells improved at

19 villages

20 additional

farmers from 5

villages identified

and trained in

Irrigation Techn.

81 heifers and 8


80 families in 8

villages

4 market started

in 4 villages

28 handpumps to

4 villages

- improvement of

Liganga Water

Tanks

20 additional

farmers from 7

villages identified

and trained in

Irrigation Techn.

Introduction of 29

Dairy Goats to 5

villages

20 heifers and 2


20 families in 5

villages.

5 markets

established in 5

villages

30 hand-pumps in

7 villages

106 Traditional

wells in 16 villages

30 additional

farmers from 4

villages identified

and trained in

Irrigation Techn.

20 heifers and 2


20 families in 5

villages

10 open markets

started

-Construction of 5

markets

proceeding.

9 hand-pumps in 3

villages

37 traditional

wells in 7 villages

90 farmers in 19

villages trained in

irrig. techniques

other farmers

adopted same

technology without

DRDP support

380 families in 42

villages supplied

with heifer/bulls

29 families supplied

with dairy goats

increased income

and nutrition

increase income and

improved supply

due to markets

69% of population

now provided with

safe water by hand

pumps and/or wells

See above


Annexes

156



HEALTH

EDUCATION

Primary Health

Care (distinct

training packages

for community

health workers,

traditional birth

attendants and

traditional healers)

Rehabilitation of

Dispensaries &

Rural Health

Centres (RHC)

Construction of

School Buildings

Integral Education

Programme

39 villages covered 40 villages covered

Run at Lusewa

division

Continuation with

villages of

previous year

6 schools were

rehabilitated in

Lusewa

20 villages covered

Mputa RHC

rehabilitated

3 schools

completed at

WIDA

- Lusew Wards

- Magazini Ward

53 classrooms

rehabilitated and

8 teacher houses

constructed

OUTPUT ACHIEVED

157


Annexes

1998 1999 2000 2001 2002 OUTCOME

6 villages covered

- Community

Health Fund

established for

Songea district

Gumbiro

Dispensary

constructed

17 classrooms

constructed &

rehabilitated

Inspection done

for 21 schools

8 schools

committee/

Teachers trained.

6 villages covered

-Community

Health Fund

inaugurated

8 Dispensaries

constructed

28 classrooms

14 Teachers House

& 28 pit latrines

constructed

- Teaching

materials

provided to 8

schools

- 62 schools

inspected

6 villages covered

3 dispensaries and

6 staff houses

constructed

21 classrooms, 11

Teachers house

constructed /

rehabilitated.

- 3 Post Primary

schools

rehabilitated

- Teaching

materials

provided to 21

classrooms

- 38 Teachers

trained

Malaria control

Programme

established

3 dispensaries & 2

staff houses

constructed

34 classrooms and

3 Teachers house

constructed

- 60 schools

inspected

- Teaching

materials

provided

- 34 classrooms

rehabilitated

- 3 posts primary

rehabilitated

2 dispensaries & 4

staff houses

constructed

54 classrooms and

20 toilets

maintained.

53 schools

inspected

172 school

committee

received training

on Basic Financial

Management.

All villages

throughout the

district covered with

training

Community Health

Fund and Malaria

Control Programme

established

Every ward has a

dispensary

See below

Until 1997 the

programme covered

only Lusewa

division, the most

neglected division in

education

From 1998 onwards

the whole district

was covered


Annexes

158



CULTURE

COOPERATIVE

Mobile

Information

Centre

Theatre Youth

Groups

Sports

Formation of

SACCOs

Women Economic

Groups

–

–

Leaders and

members of 1

SACCO trained

2 groups formed

Follow-up training

7 groups formed

17 remote villages

serviced

4 villages formed

Theatre Groups

4 villages provided

with Sports gears

3 SACCOs formed

& training of

leaders

13 groups formed

OUTPUT ACHIEVED

159


Annexes

1998 1999 2000 2001 2002 OUTCOME

3 editions of AMKA

newsletter (6000

copies) produced

& distributed

-14 Rural

Reporters trained

6 villages formed

Theatre Groups

4 villages provided

with Sports gears

1 SACCOs formed

& training of

leaders

4 Women Leaders

of 4 groups

trained

4 editions, 8,000

copies of AMKA

newsletter

produced &

distributed

4 villages formed

Theatre Groups

4 villages provided

with Sports gears

Training of 501

SACCO members

(of the 4 SACCOs

formed)

–

4 editions, 8,000

copies of AMKA

newsletter

produced &

distributed

2 villages formed

Theatre Groups

4 villages provided

with Sports gears

Follow-up training

conducted for

1,000 Primary

Society members

Further

strengthening of

the 4 SACCOs

–

3 editions, 6,000

copies of AMKA

newsletter

produced &

distributed

–

2 villages provided

with Sports gears

Consultant from

FISEDA to advise

on formation of

strong Rural

Financial

Institution

–

2 editions, 4,000

copies of AMKA

newsletter

produced &

distributed

–

2 villages provided

with Sports gears

6 Wards selected,

and Rural

Financial

Institutions

established

–

Newsletter

distributed to VEO

for sale

Whole district

covered by AMKA

newsletter

Educative plays are

performed

(HIV/AIDS, Culture

& Values)

Youth engaged in

productive activities

5 SACCOs formed

and trained

6 Rural Financial

Institutions formed

Primary Societies

strengthened

22 groups formed


Annexes

160



THIRD PARTY

FUND

NATURAL

RESOURCES

Collaboration with

NGOs / CBOs

Tree Planting

Combating

Deforestation

Bee-keeping

Fish Farming

– – - joint primary

health care pro-

gramme with

Peramiho Mission

- joint primary edu-

cation programme

and support to

SACCOs with

WIDA

- continuation

with Peramiho

- continuation

with WIDA

- collaboration

with Chamber of

Commerce on

business training

50,000 seedlings

distributed to 12

villages

125,ooo seedlings

of timber trees for

income generation

18 groups formed

in 7 villages

10 Key farmers in

5 villages provided

with fingerlet fish

for raising

OUTPUT ACHIEVED

161


Annexes

1998 1999 2000 2001 2002 OUTCOME

- continuation

with Peramiho

- continuation

with WIDA

- continuation

with Chamber of

Commerce

50,000 seedlings

distributed to 12

villages

125,ooo seedlings

of timber trees for

income generation

18 groups formed

in 7 villages

10 key farmers in 5

villages provided

with fingerlet fish

Quarterly review

meeting

conducted

46,503 seedlings

produced in 19

school nurseries

6 villages selected

to combat

deforestation

20 groups formed

in 11 villages

70 farmers in 2

villages provided

with fingerlet fish

Quarterly review

meeting

conducted

47,000 trees raised

(timber)

10 villages

produced tree

seedlings

20 bee-keepers

trained in 2

villages

113 farmers in 2

villages provided

with fingerlet fish

Quarterly review

meeting

conducted

140,967 trees

planted

10 villages

sensitised

70000 seedlings

raised

20 Bee-keepers

trained in 5

villages

57 fish farmers in

3 villages provided

with fingerlet fish

Quarterly review

meeting

conducted

140,967 trees

planted

40,000 seedlings

raised

Collaboration

between SDC and

NGO’s initiated and

continued as

discussion forum

Areas reforested

threatened by

erosion or as

identified during

VLP

As above

Increased honey

production and

income

260 farmers in 15

villages provided

with fish for own

consumption and

for selling


Annexes

162



STRENGTHENING

DISTRICT

COUNCIL

Capacity Building

District

Management

Inception of DRDP

OUTPUT ACHIEVED

163


Annexes

1998 1999 2000 2001 2002 OUTCOME

Planning Capacity

in SDC established

- Gender Task

Force established

- District Private

Sector Support

established

- 2 Car pool

system

introduced

Comprehensive

study of Council

Capacity

- Training HoDs

on Log-Frame,

linked to the

preparation of a

DDP

- introduction of

benchmark sys-

tem to streng-

then Revenue

Collection system

and Internal

Control

- review of

Organisation and

Management

arrangements

- District

Monitoring

System

introduced

- Introduction of

Gender Task

Force

- Introduction of

Computers

- Handing over

DRDP Admin.

functions. to SDC

- Training of

Councillors on

Basic Principle of

Planning,

Finance

Management

- Training for

Finance Staff

- Purchase of com-

puters, Copier,

Motorcycle.

- 27 staff trained in

computer skills

- Study visit to

Kondoa, Monduli

and Same to

HoDs and CRT

Team

- Council Reform

Programme star-

ted to implement

step No.1 – 3

- Continuing with

Handing Over

process

- Local

Government

Reform

Programme

(LGRP)

continued to Step

No.6

- DDP Focused on

Result and

Realistic Budget

- Green Sector

Study conducted.

- Refresher course

to HoDs on

Planning

- Continuing with

Handing Over

process

- Reform

continued to Step

No.6

No. 8 and 11

- Car Pool systems

integrated and

strengthened

- DDP Focused on

Result and

Realistic Budget

- Gender

Mainstreaming

Plan developed

- Green Sector

Study conducted.

- Integrated DDP

prepared

- Management

Course for HoDs

- Participatory

planning system

developed and

implemented.

- Integrated DDP

prepared and

approved

All Councillors in

district are trained

- Increased capacity

in SDC, resulting

in: SDC is natio-

nally the second

best performing

council in imple-

menting the LGPR


Annexes

164

Annex 4 References

AIV (Advisory Council on International Affairs), 2003, Pro-poor Growth in the Bilateral

Partner Countries in Sub-Saharan Africa - An Analysis of Poverty Reduction Strategies,

Report No. 29, The Hague.

Bastiaensen, J., T. de Herdt and J. Vaessen, 2002, Poverty, Institutions and Interventions.

A framework for an institutional analysis of poverty and local anti-poverty interventions,

Institute of Development Policy and Management, University of Antwerp, Antwerp.

Bigsten A., D. Mutalemwa, Y. Tsikata and S. Wangwe, 2001, Chapter 6 on Tanzania, in

S. Devarajan et al, Aid and Reform in Africa, World Bank, Washington D.C.

Blair, H., 2000, Participation and Accountability at the Periphery: Democratic Local

Governance in Six Countries, World Development, Vol. 28, No. 1, pp. 21-29.

Bol, D, 1995, The Effectiveness of Aid Intentions, Paper presented at the conference on

Aid Effectiveness in Tanzania, January 1995, Consultants for Development Programmes

(CDP) and Economic Research Bureau, Dar es Salaam.

Bol, D., N. Luvanga and J. Shitundu, eds., 1997, Economic Management in Tanzania;

Economic Research Bureau, University of Dar es Salaam, Dar es Salaam.

Bol, Dirk, 1999, District rural development projects in Tanzania, in Sterkenburg, Jan and

Arie van der Wiel, Integrated Area Development, Experiences with Netherlands Aid in

Africa, Focus on Development, No. 10, Ministry of Foreign Affairs, The Hague.

Booth, D. et al., 1993, Social, economic and cultural change in contemporary Tanzania.

A people-oriented focus, SIDA, Stockholm.

Bossuyt, Jean and Jeremy Gould, 2000, Decentralisation and Poverty Reduction:

Elaborating the Linkages, European Centre for Development Policy Management and

Institute of Development Studies, University of Helsinki, Policy Management Brief

No. 12, Maastricht/Helsinki.

165


Annexes

Carney, D., 1998, Implementing the Sustainable Rural Livelihoods Approach,

Paper presented at the 14th annual meeting between the European Fisheries Corporation

Advisors and the Commission, Goteborg.

Chachage C.S.L., B.O. Koda, Y.Q. Lawi and E.P.A.N. Mihanjo, 2001, Location Specific

Poverty Alleviation in Songea and Mbulu Districts: a research report, University of Dar es

Salaam.

CDP, 1990, Mid-term Review DRDPs, Phase I, Dar es Salaam/Utrecht.

CDP, 1994, Mid-term Evaluation District Rural development Programme in Tanzania,

Phase II 1991-1995, Dar es Salaam/Utrecht.

Conyers, Diana, Paul Mosley and Dennis M. Warren, 1988, Introduction to Special Issue

on Integrated Rural development, Manchester Papers on Development,

Vol. IV, No. 1, pp. 1-3.

Cornia A., R. Jolly and F. Stewart, 1987, Adjustment with a Human Face.

Protecting the Vulnerable and Promoting Growth, Oxford.

Coulson, A., 1982, Tanzania: a Political Economy, Clarendon Press, Oxford.

Dalgaard, C.J. and H. Hansen, 2000, On Aid, Growth and Good Policies,

Journal of Development Studies, Vol. 37, no. 6.

Doriye, J., H. White and M. Wuyts, Fungibility and Sustainability, Import support aid to

Tanzania, SIDA evaluation report 1993/5, Stockholm/The Hague.

Devarajan, S., D. Dollar and T. Holmgren (eds.), 2001, Aid and Reform in Africa,

World Bank, Washington DC.

Ellis, Frank and Godfrey Bahiigwa, 2003, Livelihoods and Rural Poverty Reduction in

Uganda, World Development, Vol. 31, No. 6, pp. 997-1013.

ETC, 2000, Impact Assessment of Activities in the Productive Sectors of Mbulu and

Karatu DRDPs, Nairobi.


Annexes

166

ETC, 2001, Study Streamlining of On-going Agricultural Projects and Programmes in the

context of the Agricultural Sector Development Strategy, Dar es Salaam.

ETC, 2001a, SWOT Analysis of the Green Sector, Songea District Council, Final Report,

Nairobi.

Gulliver, P. H., 1995, Labour migration in a rural economy – a study of the Ngoni and

Ndendeuli of southern Tanganyika, East African Institute of Social Research, Kampala.

Hyden, G., 1980, Beyond Ujamaa in Tanzania- underdevelopment and an uncaptured

peasantry, Heinemann Educational Books, London.

Iliffe, J. 1979, A Modern History of Tanganyika, Cambridge University Press, Cambridge.

ILO, 1976, Employment, Growth and Basic Needs, A One-world Problem, Geneva.

IOV, 1994, Tanzania. Evaluation of the Netherlands Development Programme with

Tanzania, 1970-1992, Operations Review Unit (IOV) of the Ministry of Foreign Affairs,

The Hague.

IOB, 2003, Co-ordination and Sector Support, An evaluation of the Netherlands’ support

to local governance in Uganda, Policy and Operations Review Unit (IOB), Ministry of

Foreign Affairs, The Hague.

Land, Tony and Volker Hauck, 2003, Building coherence between sector reforms and

decentralisation: do SWAps provide the missing link?, Discussion Paper No.49,

European Centre for Development Policy Management, Maastricht.

Luwaitama A.F. et al, 1995, Who do the Poor consider to be Poor: A critical Language

Study of conceptions of poverty by the poor in Tanzania: unpublished research report,

REPOA.

Mbulu District Council (MDC), 2001, Mbulu District Rural Development Plan, Mbulu.

Ministry of Foreign Affairs, 1991, A World of Difference, A new framework for development

co-operation in the 1990s, The Hague.

167


Annexes

Ministry of Foreign Affairs, 2001, Poverty alleviation; Dutch policy in brief, The Hague.

Ministry of Foreign Affairs, 2002, Notitie Decentralisatie en lokaal bestuur

(A Note on Decentralisation and Local Governance), Letter to Parliament, The Hague.

Netherlands Economic Institute (NEI) in association with Consultants for Water and

Environment (IWACO) and Agro-Industrial Services Ltd. (AGRIS), 1997, Towards Better

Use of Environmental Resources, a planning document for Mbulu and Karatu districts,

Tanzania, Mbulu.

OECD/DAC, 2000, DAC Guidelines on Poverty Reduction, Vol. 1 (draft), Paris.

Royal Netherlands Embassy (RNE), Annual Plan/Annual Report, 1998; 1999; 2000;2001

and 2002, Dar es Salaam.

RNE and Netherlands Development Organisation (SNV), 2000, Mid-term Review of the

District Rural Development Programme, Tanzania.

RNE and SNV, 2001, Reaction to the Mid-term Review of the District Rural Development

Programme, Dar es Salaam.

Schou, A. and J. Steffensen, 2002, Supporting decentralisation and local governance –

lessons learned, good practices and emerging issues, Report prepared by the Norwegian

Institute for Urban and Regional Research.

SDC (Songea District Council), 1998, Songea District Development Plan and Forwarding

Budget, 1999-2001, Songea.

Sen, A., 1999, Development as Freedom, Knopf, New York.

SNV, 1997, Spinning three strings into one cable, SNV’s Position Paper on the SNV sup-

ported District Rural Development Programme in Kondoa, Monduli and Songea District.

SNV, 1999, (2000 and 2002), Songea DRDP, Annual Progress Reports, Songea.


Annexes

168

SNV, 2000, SNV-supported District Rural Development Programme

Kondoa/Monduli/Songea District, Phase II, Progress and Financial Report.

Snyder, K. A., 1993, Like Water and Honey - moral ideology and the construction of com-

munity among the Iraqw of Northern Tanzania; unpublished Ph D thesis, Yale University.

Songea DRDP, 1999;2000,2002, Annual Progress Reports, Songea

Stephenson J, 2002, The Language of the Land, St. Martin’s Press, Tririon Uitgevers,

Baarn.

Sterkenburg Jan en Arie van der Wiel, (eds), 1999, Integrated Area Development-

experiences with Netherlands aid in Africa, Focus on Development No. 10, Ministry of

Foreign Affairs, The Hague.

Streeten, P et al., 1981, First Things First, Meeting basic human needs in the developing

countries, Oxford University Press, Oxford.

URT (United Republic of Tanzania), 1998, Vision 2025, Long-term development strategy,

Dar es Salaam.

URT, 1999, National Poverty Eradication Strategy, Dar es Salaam.

URT, 2000, Poverty Reduction Strategy Paper (PRSP), Government Printer, Dar es Salaam.

URT, 2001b, Rural Development Strategy (draft), Prime Minister’s Office, Dar es Salaam.

URT, 2001a, National Framework Participatory Planning and Budgeting at District Level,

Dar es Salaam.

URT, 2002, Household Budget Survey 2000/01, National Bureau of Statistics,

Dar es Salaam.

UNDP, 1996, and other years, Human Development Report, New York.

UNDP, 1997, Participatory Poverty Assessment Shinyanga Region, Dar es Salaam.

169


Annexes

Voipio T. and P. Hoebink, 1999, European Aid for Poverty Reduction in Tanzania; ODI

Working Paper 116, Institute of Development Studies (IDS), Helsinki .

World Bank, 1974, Redistribution with Growth, Washington D.C.

World Bank, 1990, Poverty, World Development Report, Washington DC.

World Bank, 1998, Assessing Aid. What works, what doesn’t, and why, Oxford University

Press, New York.

World Bank, 2000, Voices of the Poor, Vol. I Can Anyone Hear Us?

Vol. 2 Crying out for change, Vol. 3 From many lands, Oxford University Press, New York.

World Bank, 2001, Attacking Poverty, World Development Report 2000/2001,

Oxford University Press, New York.

World Bank, 2001, Country Economic Memorandum, Washington DC.


Annexes

170

Annex 5 Exchange rates 1986 –2001

171


Annexes

Exchange rates 1986 –2001

Year US$= ¤ US$= Tsh ¤ = Tsh US$ = Dfl

1986 32.7 2.45

1987 64.3 2.03

1988 99.3 1.98

1989 143.4 2.12

1990 195.1 1.89

1991 219.2 1.87

1992 297.7 1.75

1993 410.0 1.86

1994 0.828 588.2 487.0

1995 0.731 769.2 562.3

1996 0.767 769.2 590.0

1997 0.887 714.3 633.6

1998 0.901 725.7 653.9

1999 0.937 792.4 742.5

2000 1.085 738.0 800.1

2001 1.089 729.7 794.6

Evaluatie-studies uitgebracht door de inspectie ontwikkelingssamenwerking en beleidsevaluatie (iob) 1993-2003

258 1993 Evaluatie en Monitoring. De Rol van Projectevaluatie en monitoring in de Bilaterale hulp *)

259 1993 Samenwerkingsverbanden in het Hoger Onderwijs. Een evaluatie van samenwerkingsverbanden in Botswana, Lesotho, Swaziland en Tanzania

260 1994 Evaluatie van de Nederlandse Hulp aan India, Mali enTanzania. Samenvattend rapport

261 1994 India. Evaluation of the Netherlands development programme with India, 1980-1992

262 1994 Mali. Evaluatie van de Nederlandse hulp aan Mali, 1975-1992

263 1994 Tanzania. Evaluation of the Netherlands development programme with Tanzania, 1970-1992

264 1994 Humanitarian Aid to Somalia265 1995 Fertiliser Aid. Evaluation of Netherlands fertiliser aid 1975-1993

with special reference to Bangladesh, Mali and Zambia.266 1996 Netherlands Aid Reviewed. An analysis of

Operations Review Unit Reports, 1983-1994267 1997 Vrouwen in Burkina Faso en de Nederlandse

Ontwikkelingssamenwerking 1985-1995267 1997 Les Femmes du Burkina Faso et la Coopération

Néerlandaise 1985-1995268 1998 Vrouwen in Kenia en de Nederlandse

Ontwikkelingssamenwerking 1985-1995268 1998 Women in Kenya and the Netherlands Development

Cooperation 1985-1995. isbn 90-5328-152-3269 1998 Bangladesh. Evaluation of the Netherlands Development

Programme with Bangladesh, 1972-1996(Volume 1- Summary Report)

269 1998 Bangladesh. Evaluation of the Netherlands DevelopmentProgramme with Bangladesh, 1972-1996(Volume 2 - Main Report)

270 1998 Bangladesh. Evaluation of Netherlands-Funded NGO’s, 1972-1996 (Volume 3 - Sub-Report)

271 1998 Vrouwen en Ontwikkeling. Beleid en uitvoering in deNederlandse Ontwikkelingssamenwerking 1985-1996. isbn 90-5328-168-1

272 1998 SNV - Bénin, 1985 - 1995 (franstalig)272 1998 SNV - Benin, 1985 - 1995 (Hoofdbevindingen en samenvatting) *)

273 1998 SNV - Nepal, 1985 - 1995 (engelstalig). isbn 90-5328-164-9273 1998 SNV - Nepal, 1985 - 1995 (Hoofdbevindingen en samenvatting)274 1998 Evaluation of SNV in Benin, Nepal and Bolivia

(Summary evaluation report)275 1998 Egypt. Evaluation of the Netherlands Development

Programme with Egypt, 1975-1996(Volume 1 - Summary Report). isbn 90-5328-198-3

275 1998 Egypt. Evaluation of the Netherlands Development Programmewith Egypt, 1975-1996(Volume 2 - Main Report). isbn 90-5328-199-1

276 1998 Egypt. Evaluation of the Netherlands Support to WaterManagement and Drainage, 1975-1996(Volume 3 - Sub-Report). isbn 90-5328-185-1

277 1998 Bolivia. Evaluation of the Netherlands Development Programmewith Bolivia. Main Findings and summary (Volume 1 - Summary Report). isbn 90-5328-205-X

277 1998 Bolivia. Evaluation of the Netherlands Development Programmewith Bolivia. (Volume 2 - Main Report). isbn 90-5328-204-1

278 1998 The Netherlands Programme Aid to Bolivia (Volume 3 - Sub Report). isbn 90-5328-155-X

279 1999 Diamonds and Coals. Evaluation of the Matra programme ofassistance to Central and Eastern Europe,1994-1997 (Summary evaluation report). isbn 90-5328-229-7

279 1999 Diamonds and Coals. Evaluation of the Matra programme ofassistance to Central and Eastern Europe,1994-1997 (Evaluation report). isbn 90-5328-230-0

280 1999 Cofinancing between the Netherlands and the World Bank, 1975-1996 (Volume I - Summary Report). isbn 90-5328-232-7

280 1999 Cofinancing between the Netherlands and the World Bank, 1975-1996 (Volume 2 - Main Report). isbn 90-5328-231-9

281 1999 Hulp door handel, evaluatie van het Centrum tot Bevorderingvan de Import uit ontwikkelingslanden. isbn 90-5328-246-7

282 1999 Palestinian territories, review of the Netherlands developmentprogramme for the Palestinian territories, 1994-1999.isbn 90-5328-245-9

283 1999 Oret/Miliev review 1994-1999 Assisting developing countries tobuy investment goods and services in the Netherlands.isbn 90-5328-248-3

284 2000 Institutional Development Netherlands support to the water sector.isbn 90-5328-274-2

285 2000 Onderzoek naar de samenwerking tussen Mali enNederland 1994-1998isbn 90-5328-278-5

286 2001 Smallholder Dairy Support Programme (SDSP) TanzaniaInspection of its identification, formulation and tendering processisbn 90-5328-298-x

287 2001 De kunst van het Internationaal cultuurbeleid 1997-2000isbn 90-5328-300-5

288 2002 Health, nutrition and populationBurkina Faso Mozambique Yemenisbn 90-5328-301-3

289 2002 Cultuur en OntwikkelingDe evaluatie van een beleidsthema (1981-2001)isbn 90-5328-302-1

289 2002 Culture and DevelopmentEvaluation of a policy (1981-2001)isbn 90-5328-305-6

290 2003 Agenda 2000Hoe Nederland onderhandelt met Europaisbn 90-5328-307-2

291 2002 Nederlands schuldverlichtingsbeleid 1990-1999isbn 90-5328-306-4

292 2003 Resultaten van internationale schuldverlichting 1990-1999isbn 90-5328-310-2

292 2003 Results of International Debt Relief 1990-1999isbn 90-5328-314-5

293 2003 Netherlands-FAO Trust Fund Co-operation 1985-2000isbn 90-5328-308-0

294 2003 Co-ordination and Sector SupportAn evaluation of the Netherlands’ support to local governance in Uganda, 1991-2001isbn 90-5328-311-0

295 2003 Behartiging van de buitenlandse belangen van de Nederlandse Antillen en ArubaEen evaluatie van de rol van het Ministerie van Buitenlandse Zaken isbn 90-5328-316-0

296 2003 Poverty, policies and perceptions in TanzaniaAn evaluation of Dutch aid to two district rural development programmes isbn 90-5328-337-4

*) Niet meer beschikbaar

Ministry of Foreign Affairs | P.O. Box 20061 | 2500 eb The Hague | The Netherlands

Policy and Operations Evaluation Department | 2004

isbn 90-5328-337-4

Ordercode: OSDR0517/E

Documents

Poverty, policies and perceptions in Tanzania